<<

European Journal of Phycology

ISSN: 0967-0262 (Print) 1469-4433 (Online) Journal homepage: http://www.tandfonline.com/loi/tejp20

The genera Bornet & Falkenberg and Vertebrata S.F. Gray constitute well-defined clades of the red algal tribe Polysiphonieae (, )

Pilar Díaz-Tapia, Lynne McIvor, D. Wilson Freshwater, Heroen Verbruggen, Michael J. Wynne & Christine A. Maggs

To cite this article: Pilar Díaz-Tapia, Lynne McIvor, D. Wilson Freshwater, Heroen Verbruggen, Michael J. Wynne & Christine A. Maggs (2017) The genera Melanothamnus Bornet & Falkenberg and Vertebrata S.F. Gray constitute well-defined clades of the red algal tribe Polysiphonieae (Rhodomelaceae, Ceramiales), European Journal of Phycology, 52:1, 1-30, DOI: 10.1080/09670262.2016.1256436 To link to this article: http://dx.doi.org/10.1080/09670262.2016.1256436

View supplementary material Published online: 01 Feb 2017.

Submit your article to this journal Article views: 181

View related articles View Crossmark data

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=tejp20

Download by: [UNAM Ciudad Universitaria] Date: 04 May 2017, At: 14:07 EUROPEAN JOURNAL OF PHYCOLOGY, 2017 VOL. 52, NO. 1, 1–30 http://dx.doi.org/10.1080/09670262.2016.1256436

The genera Melanothamnus Bornet & Falkenberg and Vertebrata S.F. Gray constitute well-defined clades of the red algal tribe Polysiphonieae (Rhodomelaceae, Ceramiales) Pilar Díaz-Tapiaa,b,c, Lynne McIvord, D. Wilson Freshwatere, Heroen Verbruggen c, Michael J. Wynnef and Christine A. Maggs b,d aBioCost Research Group, Universidad de A Coruña, 15071, A Coruña, Spain; bFaculty of Science and Technology, Bournemouth University, Talbot Campus, Poole, Dorset BH12 5BB, UK; cSchool of BioSciences, University of Melbourne, Victoria 3010, Australia; dSchool of Biological Sciences, Queen’s University Belfast, Medical Biology Centre, BT9 7BL, Northern Ireland; eCenter for Marine Science, University of North Carolina at Wilmington, 5600 Marvin Moss Lane, Wilmington, NC 28409, USA; fDepartment of Ecology and Evolutionary Biology and Herbarium, University of Michigan Ann Arbor, MI 48109, USA

ABSTRACT is the largest genus of , and several schemes subdividing it into smaller taxa have been proposed since its original description. Most of these proposals were not generally accepted, and currently the tribe Polysiphonieae consists of the large genus Polysiphonia (190 species), the segregate genus Neosiphonia (43 species) and 13 smaller genera (< 10 species each). In this paper, phylogenetic relationships of the tribe Polysiphonieae are analysed, with particular emphasis on the genera Carradoriella, Fernandosiphonia, Melanothamnus, Neosiphonia, Polysiphonia sensu stricto, Streblocladia and Vertebrata. We evaluated the consistency of 14 selected morphological characters in the identified clades. Based on molecular phylogenetic (rbcL and 18S genes) and morphological evidence, two speciose genera are recognized: Vertebrata (including the type species of the genera Ctenosiphonia, Enelittosiphonia, Boergeseniella and ) and Melanothamnus (including the type species of the genera Fernandosiphonia and Neosiphonia). Both genera are distinguished from other members of the Polysiphonieae by synapomorphic characters, the emergence of which could have provided evolutionarily selective advantages for these two lineages. In Vertebrata trichoblast cells are multinucleate, possibly associated with the development of extraordinarily long photoprotective trichoblasts. Melanothamnus has 3-celled carpogonial branches and plastids lying exclusively on radial walls of the pericentral cells, which similarly may improve resistance to damage caused by excessive light. Other relevant characters that are constant in each genus are also shared with other clades. The evolutionary origin of the genera Melanothamnus and Vertebrata is estimated as 75.7–95.78 and 90.7–138.66 Ma, respectively. Despite arising in the Cretaceous, before the closure of the Tethys Seaway, Melanothamnus is a predominantly Indo-Pacific genus and its near-absence from the north-eastern Atlantic is enigmatic. The nomenclatural implications of this work are that 46 species are here transferred to Melanothamnus, six species are transferred to Vertebrata, and 13 names are resurrected for Vertebrata.

ARTICLE HISTORY Received 1 October 2016; Revised 28 October 2016; Accepted 30 October 2016

KEYWORDS Biogeography; evolution; molecular systematics; morphology; phylogeny; Polysiphonia; red algae; time calibration

Introduction the most speciose is the Polysiphonieae with over 300 species in 15 currently recognized genera (Guiry & The Rhodomelaceae Areschoug is the largest family Guiry, 2016). of red algae, currently including more than 1000 Within the Polysiphonieae the genus Polysiphonia species (Guiry & Guiry, 2016). It consists of the tribes Greville (1824), nom. cons., has representatives Amansieae Schmitz (1889), Bostrychieae Falkenberg throughout the world, in the majority of photic mar- (1901), Chondrieae Schmitz & Falkenberg (1897), ine benthic habitats including brackish ones (e.g. Herposiphonieae Schmitz & Falkenberg (1897), Hollenberg, 1942, 1944, 1968a, 1968b; Womersley, Heterocladieae Falkenberg (1901), Laurencieae 1979; Maggs & Hommersand, 1993; Lam et al., Schmitz (1889), Lophothalieae Schmitz & 2013). Polysiphonia is poorly circumscribed, and has Falkenberg (1897), Neotenophyceae Kraft & I.A. remained in a state of taxonomic flux since its origi- Abbott (2002), Pleurostichidieae (Hommersand, nal description. Numerous schemes for subdividing 1963), Polysiphonieae Schmitz (1889), Polyzonieae this large and morphologically diverse genus into Schmitz & Falkenberg (1897), Pterosiphonieae smaller and more manageable groups have been pro- Falkenberg (1901), Rhodomeleae Schmitz & posed (e.g. Segi, 1951; Hollenberg, 1968a, 1968b), Falkenberg (1897), Sonderelleae L.E.Phillips (2001) based mostly on the number of periaxial cells, either and Streblocladieae nom. nud. (Hommersand, 1963; four (subgenus Oligosiphonia) or more than four Kraft & Abbott, 2002; Womersley, 2003), of which

CONTACT Pilar Díaz-Tapia [email protected] © 2017 British Phycological Society 2 P. DÍAZ-TAPIA ET AL.

(subgenus Polysiphonia). These schemes have gener- relationship between Neosiphonia species and ally been rejected and several generic names (e.g. Melanothamnus somalensis, the type species of the Orcasia Kylin (1941), based on Polysiphonia senticu- genus Melanothamnus Bornet & Falkenberg, which losa Harvey) are currently regarded as synonyms of was regarded as incertae sedis (Falkenberg, 1901). Polysiphonia. However, despite having been sub- Given the taxonomic and nomenclatural complex- sumed within Polysiphonia in most classification ity within the Polysiphonieae, our aims were to re- schemes, Vertebrata S.F.Gray (1821) is currently evaluate the morphological features of Neosiphonia recognized as a monospecific genus containing only and Vertebrata in relation to those of the type species, V. lanosa. Fernandosiphonia, Streblocladia, Carradoriella, The segregate genus Neosiphonia M.-S.Kim & I.K. Melanothamnus and Polysiphonia sensu stricto within Lee (Kim & Lee, 1999) has been widely accepted and a phylogenetic analysis of the Polysiphonieae using is now the second largest in the Polysiphonieae sequences of the plastid-encoded rbcL gene and the (Guiry & Guiry, 2016). Neosiphonia (type species: N. ribosomal DNA 18S gene (SSU). We surveyed within flavimarina from Korea) is characterized by the fol- the Polysiphonieae the distribution of a striking char- lowing features: (1) thalli erect with a main axis acteristic of the ‘Polysiphonia japonica complex’, the bearing branches; (2) branches or trichoblasts formed position of plastids on radial walls of the periaxial on every segment; (3) rhizoids cut off from pericen- cells and their absence from the outer walls such that tral cells; (4) carpogonial branches 3-celled; (5) sper- nuclei are clearly visible after staining (Maggs & matangial branches formed on a branch of modified Hommersand, 1993; McIvor et al., 2001). Likewise, trichoblasts; (6) tetrasporangia in spiral arrangement we analysed the multinucleate vs. uninucleate char- (Kim & Lee, 1999). These features contrast markedly acter of trichoblast cells, which seems to be taxono- with the key characters of Polysiphonia sensu stricto, mically significant (Maggs & Hommersand, 1993). exemplified by the type species P. stricta: prostrate axes with rhizoids in open connection with pericen- tral cells; carpogonial branches 4-celled; spermatan- Materials and methods gial branches borne directly on axes; tetrasporangia in Field collections, morphological studies and straight rows (Kim et al., 2000). In addition to literature review describing the new species N. flavimarina, Kim &

Lee (1999) also transferred 11 species of Samples of Polysiphonieae (Table S1) were collected Polysiphonia to Neosiphonia, all based on material from European Atlantic coasts, New Zealand, from Korea, and there are 43 currently recognized Australia, Taiwan, Japan, Chile, USA, South Africa species (Guiry & Guiry, 2016), not all of which exhi- and Oman and processed fresh, desiccated in silica bit the six key characters of Neosiphonia listed above. gel or preserved in ethanol. Kim & Lee (1999) considered Neosiphonia (also Type material of Fernandosiphonia unilateralis was referred to as the ‘Polysiphonia japonica complex’ obtained from the Herbarium, Botanical Museum, sensu Yoon (1986)) to be related to Fernandosiphonia Göteborg, Sweden (GB) by correspondence with the Levring, which was erected for F. unilateralis from the curator. It consisted of four permanent slides, a her- Juan Fernández Islands off Chile on the basis of its barium sheet and liquid-preserved material. unilateral development of ultimate branches (Levring, Furthermore, we studied recent collections from Juan 1941) and which currently consists of three species. Fernández Islands, the type locality. We also studied They reported that Neosiphonia differed from the type material in US and TCD of several species Fernandosiphonia principally in its branching pattern, currently assigned to Neosiphonia (Table S2; herbar- the origin of spermatangial branches, and the 3-celled ium abbreviations as in Thiers, 2016) for which the key carpogonial branches. Kim & Lee (1999) did not com- morphological characteristics (Table 1) could not be ment, however, on the possible relationship between clearly ascertained from published literature, in order Fernandosiphonia and Streblocladia F.Schmitz (in to determine their correct generic assignment. For this Schmitz & Falkenberg, 1897). Hommersand (1963) purpose, we exclusively considered the descriptions and Norris (1994) compared Fernandosiphonia (tri- provided for material from type localities or near choblasts formed spirally) with Streblocladia (tricho- them. To ensure the accuracy of our interpretation of blasts borne only adaxially). Choi et al.(2001) drew the genera, our concept of them is based on material of attention to the relationship in their 18S tree between their type species obtained from their type localities. N. japonica and Polysiphonia virgata, the type species For Streblocladia, we used material of, and sequences of Carradoriella P.C.Silva (Kylin, 1956,asCarradoria; from, the type species S. glomerulata from New Silva et al., 1996), and suggested that Neosiphonia Zealand. Carradoriella (i.e. Polysiphonia virgata) was might either be subsumed into Carradoriella or be obtained from the type locality in South Africa, and resolved as a sister to it. Recent searches of DNA the type species of Vertebrata and Melanothamnus sequence databases unexpectedly showed a possible came from Ireland and Oman respectively.

Table 1. Comparison of selected morphological characteristics among the Polysiphonia sensu stricto 1 and 2, Vertebrata, Carradoriella, Streblocladia, ‘Polysiphonia’ schneideri and Melanothamnus clades. Carradoriella ‘Polysiphonia’ schneideri Feature Polysiphonia sensu stricto 1 and 2 Vertebrata clade Streblocladia clade clade Melanothamnus Thallus habit Erect; decumbent; prostrate Erect; decumbent; prostrate Erect Erect Erect; decumbent Erect; decumbent Rhizoid connection Open Cut off Cut off Cut off Cut off Cut off Pericentral cells 4 (6–8inBryocladia cuspidata, clade 2) 6–24 5–16 4–12 4–74–9 Cortication Absent Absent/Present Present Present Absent Absent/Present Plastid arrangement Scattered Scattered Scattered Scattered Scattered Radial walls Branch/trichoblast With naked segments On every segment or with naked On every segment or With naked segments With naked segments On every segment or with arrangement segments with naked naked segments segments Branches in No Yes/No Yes/No No Yes/No No trichoblast axils Trichoblast cell Uninucleate; Multinucleate; Uninucleate; Trichoblasts absent Uninucleate; Uninucleate; nuclei and pigmentation absent pigmentation absent (present) pigmentation absent pigmentation absent pigmentation absent pigmentation Branching pattern Spiral, pseudodichotomous Spiral, pseudodichotomous, Pseudodichotomous Dorsiventral, spiral, Spiral, Dorsiventral, spiral, dorsiventral pseudodichotomous pseudodichotomous pseudodichotomous Spermatangial Replacing trichoblasts (or on a trichoblast branch in On a branch of trichoblasts (replacing On a branch of Replacing trichoblasts, On a branch of On a branch of trichoblasts, branches P. devoniensis and P. kapraunii), with or without them in V. lanosa), with/without trichoblasts, with without sterile apical trichoblasts, with/ with/without sterile apical sterile apical cells sterile apical cells sterile apical cells cells without sterile apical cells cells Carpogonial branch 4-celled 4-celled 4-celled Unknown 4-celled 3-celled Cystocarp Globular; ovoid; urceolate Globular; ovoid Ovoid Ovoid Globular Globular; ovoid morphology PHYCOLOGY OF JOURNAL EUROPEAN Cells of the ostiole Similar to the cells below Similar to the cells below Similar to the cells Larger than cells below Similar to the cells below (Similar to) Larger than cells below below Tetrasporangial Straight (slightly spiral) Straight or spiral Straight or spiral Straight or spiral Straight or spiral (Straight) Spiral rows (two per segment in Ctenosiphonia) References This work, 5, 6, 9, 15, 16, 18, 22 This work, 1, 2, 3, 6, 7, 8; 16, 18, 20, This work, 16, 21 This work, 2, 3 This work, 7, 13, 18, 22 This work, 2, 3, 4, 6, 9, 10, 11, 21, 24 12, 14, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25 References: 1Abbott & Hollenberg (1976); 2Adams (1991); 3Adams (1994); 4Bustamante et al.(2013b); 5Dawes & Mathieson (2008); 6Díaz-Tapia & Bárbara (2013); 7Díaz-Tapia et al.(2013a); 8Díaz-Tapia et al.(2013b); 9Hollenberg (1942); 10Hollenberg (1968a); 11Hollenberg & Norris (1977); 12Kim & Lee (1999); 13Kim & Kim (2014); 14Kim & Kim (2016); 15Kim et al.(1994); 16Maggs & Hommersand (1993); 17Mamoozadeh & Freshwater (2011); 18Mamoozadeh & Freshwater (2012); 19Muangmai et al.(2014); 20Segi (1949); 21Stegenga et al.(1997); 22Stuercke & Freshwater (2010); 23Uwai & Masuda (1999); 24Womersley (2003); 25Yoon (1986). 3 4 P. DÍAZ-TAPIA ET AL.

Table 2. Genus Vertebrata with new combinations resulting from the present study printed in bold, followed by basionyms and taxonomic synonyms. Binomial in Vertebrata Basionym Type material Synonyms Type locality Vertebrata constricta (Womersley) Díaz-Tapia & Maggs, comb. nov. Holotype: AD A32927 Polysiphonia constricta Womersley (1979, 497–498; Southern Australian species of Polysiphonia Kangaroo I., South Australia; 21. Greville (Rhodophyta). Australian Journal of Botany, 27: 459–528) xi.1968 Vertebrata foetidissima (Cocks ex Bornet) Díaz-Tapia & Maggs, comb. nov. Lectotype (Maggs & Hommersand, Polysiphonia foetidissima Cocks ex Bornet (1892, pp. 314–315; Les algues de P. K. A. Schousboe. . ... 1993): PC 0146017 Mémoires de la Société Nationale des Sciences naturelles et Mathématiques de Cherbourg, 28: 165–376) Plymouth, England; undated Vertebrata isogona (Harvey) Díaz-Tapia & Maggs, comb. nov. Lectotype (Womersley, 1979): BM Polysiphonia isogona Harvey (in Hooker 1855, p. 231; The botany of the Antarctic voyage. . .. Reeve, 1082304 London) Blind Bay, Cook´s Straits, New Zealand; viii.1849 Vertebrata lobophoralis (N.R.Mamoozadeh & D.W.Freshwater) D.W.Freshwater, comb. nov. Holotype: US 217938 Polysiphonia lobophoralis N.R.Mamoozadeh & D.W.Freshwater (2012, pp. 331–333; Polysiphonia sensu Bocas del Toro, Panama; 6.viii.2010 lato (Ceramiales, Florideophyceae) species of Caribbean Panama including Polysiphonia lobophoralis sp. nov. and Polysiphonia nuda sp. nov. Botanica Marina, 55: 317–347) Vertebrata nigra (Hudson) Díaz-Tapia & Maggs, comb. nov. Neotype (Maggs & Hommersand, Conferva nigra Hudson (1762, p. 481; Flora anglica. . .. Prostant venales apud J. Nourse in the Strand & 1993): BM 1067621 C. Moran in Covent-Garden, London) Marsden, Durham, England; 12.vi.1971 (Hudson) Batters Vertebrata reptabunda (Suhr) Díaz-Tapia & Maggs, comb. nov. Holotype: L 955.62.97 Hutchinsia reptabunda Suhr (1831, p. 684; Beschreibung einiger neuen Algen. Flora 14: 673–687, Biarritz, Pyrénées-Atlantiques, France 709–716, 725–731) Lophosiphonia reptabunda (Suhr) Kylin

Fresh material and herbarium samples were pre- polymerase (Bioline, UK). The PCR amplification pared as squashes, either unstained or stained with followed Nam et al.(2000) and McIvor et al.(2001). aqueous aniline blue, post-fixed in 1% HCl, and About 1250 base pairs (bp) of the rbcL gene were mounted in 80% Karo corn syrup (Bestfoods Inc., amplified and the PCR fragments for sequencing NJ, USA). Permanent slide mounts were prepared as were purified using the High Pure PCR Product vouchers and deposited in: BM, MICH, SANT, WNC Purification Kit (Roche Diagnostics Ltd, Lewes, and MEL. UK), according to the manufacturer’s instructions. A systematic review was carried out to identify The PCR products were directly sequenced commer- relevant phycological literature from around the cially by MWG-Biotech, Ebersberg, Germany. world from which to assess for each species of Type material of Fernandosiphonia unilateralis had Polysiphonieae the 14 vegetative and reproductive been preserved in formalin by Levring (1941) prior to features relevant to Neosiphonia and Vertebrata. long-term storage in ethanol (A. Athanasiados, per- Nomenclatural authorities for the species men- sonal communication). At the Leiden herbarium, var- tioned in the manuscript are provided in Tables 2–5 ious protocols for retrieving DNA from formalin- and S1–S2. preserved specimens were attempted (Kirby & Reid, 2001); the most successful was to soak and wash the material repeatedly in clean sterile water, prior to DNA extraction, PCR amplification and sequencing DNA extraction using using a Chelex-100 (Biorad, This was carried out in four different laboratories Hercules, California) protocol (Goff & Moon, 1993; using different protocols as described below. Zuccarello et al., 1999). Applying a strategy for ampli- At Queen’s University Belfast, DNA was extracted fying degraded ‘ancient’ DNA (Provan et al., 2008), from fresh, silica gel-dried or ethanol-preserved primers were designed from an alignment of material using the Qiagen DNeasy Plant Mini Kit Neosiphonia harveyi and related species in order to (Qiagen GmbH, Hilden, Germany), according to the amplify 100-bp fragments. We used the primers F183 manufacturer’s instructions, or by a CTAB method, (5′ TGCAGGTGAATCTTCTACAGCT 3′) and R383 modified after Doyle & Doyle (1987). For PCR ampli- (5′ ACGTTACCAATAATTGAAGCTGTT 3′). fication, a PTC-200 DNA Engine (MJ Research Inc.) At the University of Melbourne, DNA was was used. Except for material of Fernandosiphonia extracted from silica gel-dried material following unilateralis, all PCR amplifications were carried out Saunders & McDevit (2012). PCR amplification was using rbcLFC as the forward primer, and rbcLRD as carried out for rbcL using the primers F7/RrbcStart or the reverse primer (Nam et al., 2000; McIvor et al., F57/rbcLrevNEW (Freshwater & Rueness, 1994; 2001). All reactions contained 200 µM each of dATP, Saunders & Moore, 2013) and for 18S using the dCTP, dGTP and dTTP, 0.3 µM of each primer, 2.5 primers F47 (5′ AGCCATGCAAGTGCCTGTAT 3′) TM ′ ′ mM MgCl2, and 1.6 units of Biotaq DNA and R1867 (5 CGCAGGTTCACCTACGGAAA 3 ). EUROPEAN JOURNAL OF PHYCOLOGY 5

Table 3. Genus Vertebrata with resurrected names resulting from the present study and V. lanosa printed in bold, followed by basionyms and taxonomic synonyms. Binomial in Vertebrata Basionym Type material Synonyms Type locality Vertebrata aterrima (J.D.Hooker & Harvey) Kuntze Probable syntypes: TCD 12786-8, BM Polysiphonia aterrima J.D.Hooker & Harvey 1067593-6 and BM 1067598 New Zealand Vertebrata australis (C.Agardh) Kuntze Lectotype (Parsons, 1980): PC Cladostephus australe C.Agardh Western Australia Brongniartella australis (C.Agardh) F.Schmitz Vertebrata byssoides (Goodenough & Woodward) Kuntze Lectotype (Maggs & Hommersand, 1993): BM byssoides Goodenough & Woodward Christchurch, England; 1794 (Goodenough & Woodward) F.Schmitz Vertebrata fruticulosa (Wulfen) Kuntze Lectotype (Maggs & Hommersand, 1993): Fucus fruticulosus Wulfen Wulfen (1789), pl. 16, fig. 1 Boergeseniella fruticulosa (Wulfen) Kylin Trieste [Tergestum], Adriatic Vertebrata fucoides (Hudson) Kuntze Neotype (Maggs & Hommersand, 1993): BM Conferva fucoides Hudson 807101 (Hudson) Greville Unlocalized, undated Vertebrata furcellata (C.Agardh) Kuntze Lectotype (Maggs & Hommersand, 1993): LD Hutchinsia furcellata C.Agardh 40907 (C.Agardh) Harvey Brittany, France; undated Vertebrata hypnoides (Welwitsch) Kuntze Holotype: LD Agardh’s herbarium no. 39346 Polysiphonia hypnoides Welwitsch Lisbon, Portugal Ctenosiphonia hypnoides (Welwitsch) Falkenberg Vertebrata lanosa (Linnaeus) T.A.Christensen Holotype: LINN 1274.23 Fucus lanosus Linnaeus Iceland, undated (Linnaeus) Tandy Vertebrata simulans (Harvey) Kuntze Lectotype (Maggs & Hommersand, 1993): Polysiphonia simulans Harvey BM-K Devon, England; 20.v.1831 Vertebrata stimpsonii (Harvey) Kuntze Holotype: TCD 11956 Polysiphonia stimpsonii Harvey Hakodate Bay, Japan Enelittosiphonia stimpsonii (Harvey) Kudo & Masuda

Vertebrata subulifera (C.Agardh) Kuntze Lectotype (Maggs & Hommersand, 1993): LD Hutchinsia subulifera C.Agardh 41607 Polysiphonia subulifera (C.Agardh) Harvey Venice, Italy; undated Vertebrata thuyoides (Harvey) Kuntze Lectotype (Maggs & Hommersand, 1993): Polysiphonia thuyoides Harvey TCD Boergeseniella thuyoides (Harvey) Kylin Milltown Malbay, Ireland; 1831 Vertebrata tripinnata (J.Agardh) O.Kuntze Lectotype (Díaz-Tapia et al., 2013b): LD J. Polysiphonia tripinnata J.Agardh (1842, p. 142; Algae maris Mediterranei et Adriatici, Agardh’s Herbarium 40938 observationes in diagnosin specierum et dispositionem generum. Apud Fortin, Masson et Cie, Paris) [Kuntze transferred ‘Polysiphonia tripinnata Harvey’ to Vertebrata, presumably a typographical Trieste, Italy error as the basionym is P. tripinnata J.Agardh (1842)] Vertebrata urbana (Harvey) Kuntze Probable holotype: TCD 186 Polysiphonia urbana Harvey Table Bay, Cape Province, South Africa Note: The positions of Polysiphonia paniculata in the rbcL and 18S trees are not congruent. This suggests that these two sequences, generated from samples from Chile and California, respectively (Table S1), were obtained from different species. The assignment of this species to the genus Vertebrata therefore requires further study to clarify the identity of the two published sequences. The type locality is Peru.

Reactions were performed in a total volume of 25 µl, R753 and F57-rbcLrevNEW (Freshwater & Rueness, consisting of 5 µl 5× MyTaqTM reaction buffer, 0.7 µl 1994; Saunders & Moore, 2013). The PCR products 10 µM of forward and reverse primers, 0.125 µl 1U were purified and sequenced commercially by the – µl 1 My TaqTM DNA Polymerase (Bioline), 17.475 µl sequencing service of the University of A Coruña. MilliQ® water and 1 µl template DNA. The PCR DNA extraction, amplification and sequencing at profile consisted of initial denaturation (93ºC for 3 UNCW were as described by Stuercke & Freshwater min); 35 cycles of denaturation (94ºC for 30 s), pri- (2008). mer annealing (45ºC for 30 s), and extension (74ºC for 90 s); and final extension (74ºC for 5 min). The PCR products were purified and sequenced commer- Sequence alignment and phylogenetic analysis cially by Macrogen. At A Coruña, Melanothamnus from Oman was A total of 65 rbcL and 48 18S sequences were down- extracted using the CTAB protocol (Doyle & Doyle, loaded from GenBank and 25 new rbcL and 10 new 1987) and rbcL was amplified using the primers F7- 18S sequences were generated in this study. The 6 P. DÍAZ-TAPIA ET AL.

Table 4. Genus Melanothamnus with new combinations resulting from the present study printed in bold, followed by basionyms and taxonomic synonyms. New combination (if any) Basionym Type material Synonyms Type locality, collection date Notes Melanothamnus somalensis Bornet & Falkenberg Lectotype (here designated): PC Figs 3–8 584992 Somalia Melanothamnus afaqhusainii M.Shameel Holotype: KUH-SW SAH-127 Pakistan Melanothamnus unilateralis (Levring) Díaz-Tapia Holotype: GB Figs 9–18 & Maggs, comb. nov. Juan Fernández Islands, Chile Fernandosiphonia unilateralis Levring (1941, pp. 660–662; Die Meeresalgen der Juan Fernandez- Inseln. Die Corallinaceen der Juan Fernandez- Inseln. In: The natural history of Juan Fernandez and Easter Island (Skottsberg, C., editor) Vol. 2, 601–670; 753–757. Almqvist & Wiksells Boktryckeri, Uppsala) Melanothamnus apiculatus (Hollenberg) Díaz- Holotype: US 48522 3-celled carpogonial branches (Kim & Abbott, 2006); Tapia & Maggs, comb. nov. O’ahu Island, Hawai’i; 30.vii.1959 plastid character (Hollenberg 1968a; fig. 9) Polysiphonia apiculata Hollenberg (1968a, p. 61; An account of the species of Polysiphonia of the central and western tropical Pacific ocean. II Oligosiphonia. Pacific Science, 22: 56–98) Neosiphonia apiculata (Hollenberg) Masuda & Kogame Melanothamnus bajacali (Hollenberg) Díaz-Tapia Holotype: AHFH ‘Cell walls hyaline’ (Hollenberg, 1961 ). Molecular & Maggs, comb. nov. Isla Guadalupe, Baja California, data available from Yucatan, Mexico Polysiphonia bajacali Hollenberg (1961, pp. 347–348; Mexico; 18.xii. 1949 (Mamoozadeh & Freshwater, 2011) Marine red algae of Pacific Mexico, Part 5: The genus Polysiphonia. Pacific Naturalist, 2: 345–375) Neosiphonia bajacali (Hollenberg) N.R.Mamoozadeh & D.W. Freshwater Melanothamnus balianus (D.E.Bustamante, B.Y. Holotype: CUK 7937 Molecular data from the type locality (Bustamante Won & T.O.Cho) Díaz-Tapia & Maggs, com. Blue Lagoon beach, Padang Bai, et al., 2013b) nov. Karangasem, Bali, Indonesia; Neosiphonia baliana D.E.Bustamante, B.Y.Won & T. 27.iv. 2012. O.Cho (2013b, pp. 516–518; Neosiphonia baliana sp. nov. and N. silvae sp. nov. (Rhodomelaceae,

Rhodophyta) from Bali, Indonesia. Botanica Marina, 56: 515–524) Melanothamnus blandii (Harvey) Díaz-Tapia & Lectotype (Womersley, 1979): 3-celled carpogonial branches; plastid character. Maggs, comb. nov. TCD Molecular data available from the type locality Polysiphonia blandii Harvey (1862, pl. 184; Brighton, Port Phillip, Victoria, (this work) Phycologia australica. . .. Vol. 4. Lovell Reeve & Co, Australia London) Melanothamnus cheloniae (Hollenberg & J.N. Holotype: US 160602 Plastid character (Hollenberg & Norris, 1977; fig. 4B) Norris) Díaz-Tapia & Maggs, comb. nov. Sonora, Gulf of California, Polysiphonia sphaerocarpa var. cheloniae Hollenberg Mexico; 21.i.1974 & J.N.Norris (1977,p.16–17; The red alga Polysiphonia (Rhodomelaceae) in the Northern Gulf of California. Smithsonian Contributions to the Marine Sciences,1:1–21) Neosiphonia cheloniae (Hollenberg & J.N.Norris) J.N. Norris Melanothamnus collabens (C.Agardh) Díaz-Tapia Syntypes: LD Agardh herbarium 3-celled carpogonial branches; plastid character; & Maggs, comb. nov. 40885–40887 and molecular data available from the type locality Hutchinsia collabens C.Agardh (1824, p. 153; Systema 40890–40898; (Díaz-Tapia & Bárbara, 2013) algarum. Berlinginiana, Lundae) Cádiz, Spain Polysiphonia collabens (C.Agardh) Kützing Streblocladia collabens (C.Agardh) Falkenberg Neosiphonia collabens (C.Agardh) Díaz-Tapia & Bárbara Melanothamnus concinnus (Hollenberg) Díaz- Holotype: US 61210; Plastid character observed in our study of the type Tapia & Maggs, comb. nov. La Jolla, California, USA; 26. material Polysiphonia concinna Hollenberg (1944, pp. xii.1936 474–475; An account of the species of Polysiphonia on the Pacific coast of North America. II. Polysiphonia. American Journal of Botany, 31: 474–483) Polysiphonia johnstonii var. concinna (Hollenberg) Hollenberg Neosiphonia concinna (Hollenberg) J.N.Norris (Continued) EUROPEAN JOURNAL OF PHYCOLOGY 7

Table 4. (Continued). New combination (if any) Basionym Type material Synonyms Type locality, collection date Notes Melanothamnus decumbens (T.Segi) Díaz-Tapia & Holotype: SAP 25880; 3-celled carpogonial branches; plastid character Maggs, comb. nov. Mihonoseki, Shimane Prefecture, (Kim, 2003, fig. 5F). Molecular data available from Polysiphonia decumbens T.Segi (1951, p. 218; Japan; vi.1948 Korea (Kim & Yang, 2006) Systematic study of the genus Polysiphonia from Japan and its vicinity. Journal of the Faculty of Fisheries, Prefectual University of Mie, 1: 169–272) Neosiphonia decumbens (T.Segi) M.-S.Kim & I.K.Lee Melanothamnus ecorticatus (R.E.Norris) Díaz- Holotype: BISH 630042 Plastid character; ostiolar cells larger than other Tapia & Maggs, comb. nov. Keokea Bay, Hawai´i; v.1990 pericarpial cells (Abbott, 1999) Fernandosiphonia ecorticata R.E.Norris (1994, p. 434; Some cumophytic Rhodomelaceae (Rhodophyta) occuring in Hawaiian surf. Phycologia 33: 434–443) Melanothamnus eastwoodiae (Setchell & N.L. Holotype: CAS 173674 Plastid character observed in type material Gardner) Díaz-Tapia & Maggs, comb. nov. Islas Revillagigedo Polysiphonia eastwoodiae Setchell & N.L.Gardner (1930, p. 161, as P. eastwoodae; Marine algae of the Revillagigedo Islands expedition in 1925. Proceedings of the California Academy of Sciences, 4: 109–215) Neosiphonia eastwoodae (Setchell & N.L.Gardner) Xiang Si-duan Melanothamnus ferulaceus (Suhr ex J.Agardh) Type materials are in LD, Plastid character in Panama and Brazil. Molecular Díaz-Tapia & Maggs, comb. nov. J.Agardh´s Herbarium (not data from Panama (Guimarâes et al., 2004; Polysiphonia ferulacea Suhr ex J.Agardh (1863, p. 980; seen) Mamoozadeh & Freshwater, 2012) Species Genera et Ordines Algarum. . ... C.W.K. Atlantic Mexico; North America; Gleerup, Lundae) Guadeloupe; Hawaiian Islands; Neosiphonia ferulacea (Suhr ex J.Agardh) S.M. Marquesas Islands; Australia Guimarães & M.T.Fujii Melanothamnus fibrillosus (Okamura) Díaz-Tapia Lectotype (Uwai & Masuda, 3-celled carpogonial branches, plastid character, cells & Maggs, comb. nov. 1999): SAP surrounding the ostiole much larger than the cells Pterosiphonia fibrillosa Okamura (1912, p. 172; Icones Shirahama, Chiba Prefecture, below (Uwai & Masuda, 1999, figs 18, 19). of Japanese Algae. Vol. II (10). Privately published, Japan; undated Tokyo) Kintarosiphonia fibrillosa (Okamura) S. Uwai & Masuda

Melanothamnus flavimarinus (M.-S.Kim & I.K.Lee) Holotype: SNU IBA001 3-celled carpogonial branches; plastid character (Kim Díaz-Tapia & Maggs, comb. nov. Bangpo, Anmyondo, Korea; 17. & Lee, 1999, fig. 5). Molecular data available from Neosiphonia flavimarina M.-S.Kim & I.K.Lee (1999, vii.1988. the type locality (Kim & Yang, 2006) p. 272; Neosiphonia flavimarina gen. et sp. nov. with a taxonomic reassessment of the genus Polysiphonia (Rhodomelaceae, Rhodophyta). Phycological Research, 47: 271–281) Melanothamnus forfex (Harvey) Díaz-Tapia & Lectotype (Womersley, 1979): 3-celled carpogonial branches; plastid character, Maggs, comb. nov. TCD 15353-4 molecular data available from the type locality Polysiphonia forfex Harvey (1859, pl. XCVI; Rottnest Island, Western (this work) Phycologia Australica. . .. Vol. 2. Lovell Reeve & Co, Australia London) Melanothamnus gorgoniae (Harvey) Díaz-Tapia & Syntypes: TCD 12801-4, NY 3-celled carpogonial branches observed in Brazil Maggs, comb. nov. 900637-8 (Guimaraes et al., 2004); plastid character observed Polysiphonia gorgoniae Harvey (1853, p. 39; Nereis Key West, Florida, USA in type material (this work) boreali-americana. . .. Smithsonian Contributions to Knowledge,5:[i–ii], [1]–258, pls XIII–XXXVI) Neosiphonia gorgoniae (Harvey) S.M.Guimarães & M. T.Fujii Melanothamnus harlandii (Harvey) Díaz-Tapia & Probable syntypes: TCD 11955, 3-celled carpogonial branches. Molecular data Maggs, comb. nov. US 56848 available from Korea (Kim, 2003; Kim & Yang, Polysiphonia harlandii Harvey (1860, p. 330; Hong Kong 2006) Characters of new algae. . .. Proceedings of the American Academy of Arts and Sciences,4: 327–335) Neosiphonia harlandii (Harvey) M.S.Kim & I.K.Lee Melanothamnus harveyi (Bailey) Díaz-Tapia & Lectotype (Maggs & 3-celled carpogonial branches (this work); plastid Maggs, comb. nov. Hommersand, 1993): TCD character. Molecular data available from the type Polysiphonia harveyi Bailey (1848, p. 38; 12810 locality (McIvor et al., 2001) Continuation of the list of localities of algae in the Bailey; Stonington, Connecticut, United States. Proceedings of the American USA; vii.1847 Academy of Arts and Sciences, 4: 327–335) Neosiphonia harveyi (Bailey) M.-S.Kim, H.-G.Choi, Guiry & G.W.Saunders (Continued) 8 P. DÍAZ-TAPIA ET AL.

Table 4. (Continued). New combination (if any) Basionym Type material Synonyms Type locality, collection date Notes Melanothamnus hawaiiensis (Hollenberg) Díaz- Holotype: US 48524 3-celled carpogonial branches (Kim & Abbott, 2006). Tapia & Maggs, comb. nov. Waikiki beach, O´ahu Island, Plastid character (Abbott, 1999, fig. 122C) Polysiphonia hawaiiensis Hollenberg (1968a, pp. Hawai´i; 21.i.1963 66–67; An account of the species of Polysiphonia of the central and western tropical Pacific ocean. II Oligosiphonia. Pacific Science, 22: 56–98) Neosiphonia hawaiiensis (Hollenberg) M.-S.Kim & I. A.Abbott Melanothamnus incomptus (Harvey) Díaz-Tapia & Probable holotype: TCD 192 Plastid character. Molecular data available from the Maggs, comb. nov. False Bay, Cape Province, South type locality (this work) Polysiphonia incompta Harvey (1847, p. 44; Nereis Africa australis. . .. Reeve Brothers, London) Melanothamnus japonicus (Harvey) Díaz-Tapia & Lectotype (Masuda et al., 1995): Plastid character (this work); 3-celled carpogonial Maggs, comb. nov. TCD 11905 branches. Molecular data available from the type Polysiphonia japonica Harvey (in M.C. Perry 1857, p. Hakodate, Japan; v.1854 locality (Kim & Yang, 2006) 331; Account of the Botanical specimens. (Gray, A., editor) Narrative of the expedition of an American squadron to the China Seas and Japan. . .. Senate of the Thirty-third Congress, Second Session, Executive Document. House of Representatives, Washington) Neosiphonia japonica (Harvey) M.S.Kim & I.K.Lee Melanothamnus johnstonii (Setchell & Gardner) Holotype: CAS1361 Plastid character. Molecular data available from the Díaz-Tapia & Maggs, comb. nov. San Esteban Island, Gulf of type locality (this work). Polysiphonia johnstonii Setchell & Gardner (1924,p. California; iv.1921 The sequence from California (KX756670) is only 767; XXIX Expedition of the California Academy 0.1–0.2% divergent in its rbcL sequence from M. of Sciences to the Gulf of California in 1921. The collabens from Spain. Further studies are needed to Marine Algae. Proceeding of the California clarify the possible synonymy between these two Academy of Science, Series 4 12: 695–949) species that share the unusual character of having Neosiphonia johnstonii (Setchell & N.L.Gardner) J.N. (5–) 6 pericentral cells Norris Melanothamnus nudus (N.R.Mamoozadeh & D.W. Holotype: US 211334 Molecular data available from the type locality Freshwater) D.W.Freshwater, comb. nov. Parque de Juventud, Calle (Mamoozadeh & Freshwater, 2012) Polysiphonia nuda N.R.Mamoozadeh & D.W. Primero, Colon, Caribbean Freshwater (2012, p. 335; Polysiphonia sensu lato coast of Panama; 20.v.2009

(Ceramiales, Florideophyceae) species of Caribbean Panama including Polysiphonia lobophoralis sp. nov. and Polysiphonia nuda sp. nov. Botanica Marina, 55: 317–347) Melanothamnus peruviensis (D.E.Bustamante, B.Y. Holotype: CUK 7976 Plastid character. Molecular data available from the Won, M.E.Ramirez & T.O.Cho) Díaz-Tapia & Lagunillas, Pisco, Ica, southern type locality (Bustamante et al., 2012, fig. 10) Maggs, comb. nov. coast of Lima, Peru; 21. Neosiphonia peruviensis D.E.Bustamante, B.Y.Won, viii.2008 M.E.Ramirez & T.O.Cho (2012, p. 360; Neosiphonia peruviensis sp. nov. (Rhodomelacea, Rhodophyta) from the Pacific coast of South America. Botanica Marina, 55: 359–366) Melanothamnus platycarpus (Børgesen) Díaz-Tapia Probable syntypes: BM 1067681 3-celled carpogonial branches; plastid character & Maggs, comb. nov. and 106760 (Iyengar & Balakrishnan, 1949, fig. 1) Polysiphonia platycarpa Børgesen (1934, p. 23; Some Bombay, India; 19.xii.1927 Indian Rhodophyceae especially from the shores of the Presidency of Bombay-IV. Bulletin of Miscellaneous Information, Royal Botanic Gardens, Kew, 1934: 1–30) Melanothamnus pseudovillum (Hollenberg) Díaz- Holotype: US 61232; Cell walls ‘hyaline’ (Hollenberg, 1968a). Molecular Tapia & Maggs, comb. nov. North Island, Johnston Islands; data available from Panama (Mamoozadeh & Polysiphonia pseudovillum Hollenberg (1968a , p. 73; 22.vi.1965 Freshwater, 2011) An account of the species of Polysiphonia of the central and western tropical Pacific ocean. II Oligosiphonia. Pacific Science, 22: 56–98) Melanothamnus ramireziae (D.E.Bustamante, B.Y. Holotype: CUK 6511 Plastid character, 3-celled carpogonial branches. Won & T.O.Cho) Díaz-Tapia & Maggs, comb. Lagunillas, Pisco, Ica, Peru; 21. Molecular data available from the type locality nov. viii.2008 (Bustamante et al., 2013a, fig. 1f) Neosiphonia ramirezii D.E.Bustamante, B.Y.Won & T.O.Cho (2013a, Neosiphonia ramirezii sp. nov. (Rhodomelaceae, Rhodophyta) from Peru. Algae, 28: 73–82) Melanothamnus silvae (D.E.Bustamante, B.Y.Won Holotype: CUK 7976 Plastid character. Molecular data available from the & T.O.Cho) Díaz-Tapia & Maggs, comb. nov. Geger, Nusadua, Bali, Indonesia; type locality (Bustamante et al., 2013b, figs 22–23) Neosiphonia silvae D.E.Bustamante, B.Y.Won & T.O. 26.iv.2012 Cho (2013b, pp. 518–520; Neosiphonia baliana sp. nov. and N. silvae sp. nov. (Rhodomelaceae, Rhodophyta) from Bali, Indonesia. Botanica Marina, 56: 515–524) (Continued) EUROPEAN JOURNAL OF PHYCOLOGY 9

Table 4. (Continued). New combination (if any) Basionym Type material Synonyms Type locality, collection date Notes Melanothamnus simplex (Hollenberg) Díaz-Tapia Holotype: US 61238 Plastid character. Molecular data available from the & Maggs, comb. nov. Laguna Beach, Orange County, type locality (this work). Polysiphonia simplex Hollenberg, (1942, p. 782; An California, USA; 14.v.1937 RbcL sequence not incuded in our phylogeny because account of the species of Polysiphonia on the it is only 1% divergent from N. ramirezii Pacific coast of North America. I. Oligosiphonia. American Journal of Botany, 29: 772–785) Neosiphonia simplex (Hollenberg) Y.-P.Lee Melanothamnus sphaerocarpus (Børgesen) Díaz- Isotypes: US, C Plastid character. Molecular data available from Tapia & Maggs, comb. nov. Store Nordsidebugt, St. Thomas, Florida (Mamoozadeh & Freshwater, 2011, fig. 18) Polysiphonia sphaerocarpa Børgesen (1918, p. 271; Virgin Islands The marine algae of the Danish West Indies. Part 3. Rhodophyceae (4). Dansk Botanisk Arkiv,3: 241–304) Neosiphonia sphaerocarpa (Børgesen) M.-S.Kim & I. K.Lee Melanothamnus strictissimus (J.D.Hooker & Probable syntype: BM 561312 Plastid character (this work). Molecular data Harvey) Díaz-Tapia & Maggs, comb. nov. New Zealand available from the type locality (Stuercke & Polysiphonia strictissima J.D.Hooker & Harvey (1845, Freshwater, 2010) p. 538; Algae Novae Zelandiae. . .. London Journal of Botany, 4: 521–551) Melanothamnus thailandicus (N.Muangmai & C. Holotype: KUMF-SRC 03-011-1 3-celled carpogonial branches. Molecular data Kaewsuralikhit) Díaz-Tapia & Maggs, comb. Sri Racha Harbor, Chon Buri, available from the type locality (Muangmai et al., nov. Thailand; 11.iii.2011 2014) Neosiphonia thailandica N.Muangmai & C. Kaewsuralikhit (2014, pp. 460–461; The new species Neosiphonia thailandica sp. nov. (Rhodomelaceae, Rhodophyta) from the Gulf of Thailand. Botanica Marina, 57: 459–467) Melanothamnus yongpilii (B.Kim & M.-S.Kim) Holotype: JNUB 140704-101 3-celled carpogonial branches; plastid character. Díaz-Tapia & Maggs, comb. nov. Jongdal, Jeju Island, Korea; 04. Molecular data available from the type locality Neosiphonia yongpilii B.Kim & M.-S.Kim (2016, pp. iv.2014 (Kim & Kim, 2016, fig. 14) 324–325; Neosiphonia yongpilii sp. nov. (Rhodomelaceae, Rhodophyta), known as Neosiphonia simplex from Korea, with an emphasis on cystocarp development. Phycologia, 55: –

323 332) Note: Here we include only species that we can confidently assign to this genus (i.e. we have examined type material or suitable pictures of the type material showing the plastid character and/or sequences are available from the type locality or nearby). sequences and their corresponding GenBank acces- (1846) were selected as outgroups for the 18S analy- sion numbers are listed in Table S1. sis. This outgroup selection was based on our phylo- Sequences were aligned using Muscle in Geneious genomic analyses of the major lineages of the 6.1.8 (Kearse et al., 2012). Identical sequences and Rhodomelaceae which resolve a clade formed by the those that diverged by less than 1.1% were removed Herposiphonieae and Pterosiphonieae as sister to the from the rbcL analysis, except for Neosiphonia flavi- Polysiphonieae (Díaz-Tapia et al., 2015). marina and N. harveyi (0.4% divergence), the two We used MrBayes v.3.2.2 for Bayesian phyloge- selected representatives of the ‘N. japonica complex’ netic inference (Ronquist et al., 2011). The rbcL align- (Kim & Lee, 1999), which also includes N. decum- ment was analysed using a single (unpartitioned) bens, N. harlandii and P. akkeshiensis Segi (McIvor GTR+Γ+I as well as completely unlinked GTR+Γ+I et al., 2001; Kim & Yang, 2006; Bárbara et al., 2013; for each codon position. We used a single GTR+Γ+I Savoie & Saunders, 2015). Identical sequences were model for 18S. All analyses were run for 5 million also removed from the 18S analysis. The sequences generations, sampling every 1000th generation and included in the final alignment were selected after using two independent runs each consisting of four considering their quality in terms of both length incrementally heated Metropolis-coupled (MCMC) and the presence of ambiguous bases. Phylogenetic chains. Convergence and stationarity of runs were trees for rbcL and 18S were estimated with Maximum evaluated with Tracer v.1.6.0 (Rambaut et al., 2013), likelihood (ML) using RAxML 8.1.6 (Stamatakis, resulting in the use of a burn-in of 500k generations 2014). GTR-Gamma was selected as the best nucleo- for all analyses. Post-burn-in trees were summarized tide model; branch support was estimated with 100 with the sump command in MrBayes, using the all- bootstrap replicates. Three species of Symphyocladia compatible-groups consensus type. Falkenberg (in Schmitz & Falkenberg, 1897) were Trees were calibrated in geological time using selected as the outgroup in the rbcL phylogeny and relaxed molecular clock analyses. The calibration one species each of Symphyocladia, Xiphosiphonia was derived from node ages inferred by Yang et al. Savoie & Saunders (2016) and Herposiphonia Nägeli (2016), which estimated the earliest split in 10 P. DÍAZ-TAPIA ET AL.

Table 5. New combinations in Melanothamnus made for formal reasons (because the current genus is here placed in synonymy with Melanothamnus) although type material has not been examined. They are printed in bold, followed by basionyms and taxonomic synonyms. Binomial in Melanothamnus (if any) Basionym Type material Synonyms Type locality Notes Melanothamnus hancockii (E.Y.Dawson) Díaz-Tapia & Holotype: LAM EYD629c Plastid character. Molecular data from Japan (this work) Maggs, comb. nov. Baja California, Mexico; Polysiphonia hancockii E.Y.Dawson (1944, pp. 331–332; 16.ii.1940 The marine algae of the Gulf of California. Allan Hancock Pacific Expeditions 3: 189–432) Fernandosiphonia hancockii (E.Y.Dawson) R.E.Norris Melanothamnus masonii (Setchell & N.L.Gardner) Holotype: CAS 173618 Díaz-Tapia & Maggs, comb. nov. Isla Guadalupe, México; Polysiphonia masonii Setchell & Gardner (1930, p. 160; iv.1925 Marine algae of the Revillagigedo Islands expedition in 1925. Proceedings of the California Academy of Sciences, 4: 109–215) Neosiphonia masonii (Setchell & N.L.Gardner) J.N.Norris Melanothamnus minutissimus (Hollenberg) Díaz-Tapia Holotype: US 66797 Plastid character. Molecular data from Japan (this work) & Maggs, comb. nov. Punta Banda, Baja Polysiphonia minutissima Hollenberg (1942, p. 781, An California, Mexico; 17. account of the species of Polysiphonia on the Pacific xii.1938 coast of North America. I. Oligosiphonia. American Journal of Botany, 29: 772–785) Melanothamnus nanus (A.J.K.Millar) Díaz-Tapia & Holotype: MELU AM752 Maggs, comb. nov Coffs Harbour, New Fernandosiphonia nana A.J.K.Millar (1990, p. 439; South Wales; 27. Marine red algae of the Coffs Harbour region, viii.1980 northern New South Wales. Australian Systematic Botany, 3: 293–593) Melanothamnus notoensis (Segi) Díaz-Tapia & Maggs, Holotype: SAP 025894 Plastid character in Korea (Nam & Kang, 2012; fig. 47E) comb. nov. Shibagaki, Ishikawa Polysiphonia notoensis Segi (1951, p. 266; Systematic Prefecture, Japan; 9. study of the genus Polysiphonia from Japan and its vii.1947 vicinity. Journal of the Faculty of Fisheries, Prefectual University of Mie, 1: 169–272) Neosiphonia notoensis (Segi) M.S.Kim & I.K.Lee Melanothamnus polyphysus (Kützing) Díaz-Tapia & Holotype: L 4082747

Maggs, comb. nov. Vieillard; New Caledonia; Polysiphonia polyphysa Kützing (1863, p. 20; Tabulae undated phycologicae; oder, Abbildungen der Tange. Vol. XIII. Gedruckt auf kosten des Verfassers (in commission bei W. Köhne), Nordhausen) Neosiphonia polyphysa (Kützing) Skelton & G.R.South Melanothamnus porrectus (T.Segi) Díaz-Tapia & Holotype: SAP 025867 Plastid character in Korea (Lee, 2008, p. 314, fig. C) Maggs, comb. nov. Henashi, Nishitsugaru- Polysiphonia porrecta Segi (1951, p. 260; Systematic study gun, Aomori of the genus Polysiphonia from Japan and its vicinity. Prefecture, Japan; 19. Journal of the Faculty of Fisheries, Prefectual University vi.1948. of Mie, 1: 169–272) Neosiphonia porrecta (Segi) Y.-P. Lee Melanothamnus savatieri (Hariot) Díaz-Tapia & Lectotype (Kim, 2005): PC 3-celled carpogonial branches in Malaysia (Masuda Maggs, comb. nov. 0011879 et al., 2001). Molecular data available from Korea Polysiphonia savatieri Hariot (1891, p. 226; Liste des Yokosuka, Kanagawa (Phillips et al., 2000) algues marines rapportés de Yokoska (Japon) par M. le Prefecture, Japan Dr Savatier. Mémoires de la Société nationale des sciences naturelles de Cherbourg, 27: 211–230) Neosiphonia savatieri (Hariot) M.S.Kim & I.K.Lee Melanothamnus sparsus (Setchell) Díaz-Tapia & Holotype: UC 261144; Plastid character in Vietnam (Abbott et al., 2002; fig. 27) Maggs, comb. nov. Arue Reef, Tahiti; 27. Lophosiphonia sparsa Setchell (1926, p. 103; Tahitian vi.1922 algae collected by W.A. Setchell, C.B. Setchell and H.E. Parks. University of California Publications in Botany, 12: 61–142) Polysiphonia sparsa (Setchell) Hollenberg Neosiphonia sparsa (Setchell) I.A.Abbott Melanothamnus teradomariensis (M.Noda) Díaz-Tapia Holotype: Herbarium Molecular data available from Korea (Bárbara et al., & Maggs, comb. nov. Niigata University 2013) Polysiphonia teradomariensis M.Noda (in Noda, M. & Echigo Province, Japan; Kitami, T. 1971, 47; Some species of marine algae from 28.xi.1968 Echigo Province facing the Japan Sea. Scientific Reports Niigata University, Ser. D. (Biology),8:35–52) Polysiphonia japonica var. teradomariensis (M.Noda) H. Y.Yoon Neosiphonia teradomariensis (M.Noda) M.-S.Kim & I.K. Lee (Continued) EUROPEAN JOURNAL OF PHYCOLOGY 11

Table 5. (Continued). Binomial in Melanothamnus (if any) Basionym Type material Synonyms Type locality Notes Melanothamnus tongatensis (Harvey ex Kützing) Díaz- Holotype: L 4083619 Molecular data available from Panama (Mamoozadeh & Tapia & Maggs, comb. nov. Tonga, Friendly Islands; Freshwater, 2011) Polysiphonia tongatensis Harvey ex Kützing, (1864, p. 14; undated Tabulae phycologicae; oder, Abbildungen der Tange. Vol. XIV. Gedruckt auf kosten des Verfassers (in commission bei W. Köhne), Nordhausen) Neosiphonia tongatensis (Harvey ex Kützing) M.-S.Kim & I.K.Lee Melanothamnus upolensis (Grunow) Díaz-Tapia & Syntypes: W Molecular data available from Hawaii (Sherwood et al., Maggs, comb. nov. Upolu, Western Samoa 2010) Polysiphonia upolensis Grunow (1874, p. 49; Algen der Fidschi-, Tonga- und Samoa-Inseln, gesammelt von Dr. E. Graeffe. Journal des Museum Godeffroy,3: 23–50) Neosiphonia upolensis (Grunow) M.S.Kim & Boo Melanothamnus yendoi (T.Segi) Díaz-Tapia & Maggs, Holotype: SAP 0258883 Molecular data available from Korea (Bárbara et al., comb. nov. Muroran, Hokkaido, 2013) Polysiphonia yendoi Segi (1951, p. 211; Systematic study Japan; 30.iv.1935 of the genus Polysiphonia from Japan and its vicinity. Journal of the Faculty of Fisheries, Prefectual University of Mie, 1: 169–272) Neosiphonia yendoi (Segi) M.-S.Kim & I.K.Lee Neosiphonia saccorhiza (F.S.Collins & Hervey) J.M.C. Isotypes: NY, Collins Transfer to Melanothamnus is not made here as N. Nunes & S.M.Guimarães, nom. inval. Herbarium saccorhiza is an invalid combination (the basionym Lophosiphonia saccorhiza F.S.Collins & Hervey Gibbet Island, Bermuda was not cited), and the phylogenetic affinities of Polysiphonia saccorhiza (F.S.Collins & Hervey) Lophosiphonia saccorhiza are unknown. Hollenberg

Ceramiales (between Spyridia Harvey (1833) and the rbcL sequence with seven ambiguous nucleotides. The remaining Ceramiales) to be 292 Ma old (st dev ≈ sequence was unique by comparison with other taxa 24.6 Ma). After adding the rbcL sequences of sequenced either at QUB or in Leiden, confirming Ceramiales from the Yang et al.(2016) study to our that there had been no contamination. alignment and setting Spyridia as the outgroup, node Twenty-five new rbcL sequences and ten 18S ages were inferred with two Bayesian methods. The sequences were obtained from members of the first analysis used an autocorrelated model of mole- Polysiphoniae (Table S1), including an rbcL sequence cular evolutionary rate change (Thorne & Kishino, from Melanothamus somalensis, and four rbcL and 2002) as implemented in PhyloBayes v.3.3f (Lartillot two 18S sequences from new collections of F. unila- et al., 2009). The MCMC chain was run for 50k teralis from the type locality. Alignments for the rbcL cycles, stationarity was assessed with Tracer, and the were unambiguous, with no insertions or deletions. node ages summarized with the readdiv command, discarding the first 25k cycles as burn-in. The second analysis used uncorrelated rates of evolution sampled Phylogenetic analyses from a lognormal distribution (Drummond et al., 2006) as implemented in BEAST v.1.8.2. The The ML rbcL tree (Fig. 1) has three strongly sup- MCMC chain was run for 10 million generations, ported major clades within the Polysiphonieae: used a Yule tree prior, and an unpartitioned GTR+Γ Polysiphonia sensu stricto 1 (including P. stricta, the +I model of sequence evolution. Stationarity was type of the genus), Polysiphonia sensu stricto 2 (with assessed with Tracer. A maximum clade credibility morphological features corresponding to those defin- tree and median node heights were inferred with ing Polysiphonia sensu stricto: Kim et al., 2000) and a TreeAnnotator, discarding the first 1 million genera- third clade grouping all the other taxa. The third tions as burn-in and using a posterior probability clade comprised a large number of lineages, many limit of zero. with low or intermediate support. The two most speciose lineages, here named Vertebrata and Melanothamnus, however, are both robustly sup- Results ported (Fig. 1). The Vertebrata clade includes V. lanosa, the current name for the type species of DNA sequences and alignments Vertebrata, V. fastigiata S.F.Gray (1821), as well as DNA extraction and PCR amplification of type mate- the type species of several other genera: rial of Fernandosiphonia unilateralis that had been Brongniartella Bory (1822), Boergeseniella Kylin initially formalin/seawater fixed and then stored in (1956), Enelittosiphonia Segi (1949) and ethanol for several decades yielded a 95 bp partial Ctenosiphonia Falkenberg (in Schmitz & Falkenberg, 12 P. DÍAZ-TAPIA ET AL.

Fig. 1. Phylogenetic tree estimated with ML analysis of rbcL sequences. Values at nodes indicate bootstrap support (BP)/ posterior probability (PP) (only shown if > 60/0.6). Branches marked with an asterisk received 100% (BP)/1.00 (PP) support. Species names printed in bold correspond to type species of genera. EUROPEAN JOURNAL OF PHYCOLOGY 13

1897). The Melanothamnus clade includes and Streblocladia clades, in which all species are erect. Fernandosiphonia unilateralis, Neosiphonia flavimar- True prostrate axes giving rise to erect axes, as opposed ina and M. somalensis, the type species of their cor- to decumbent axes that themselves become erect, are responding genera. In addition to these two large confined to Polysiphonia sensu stricto and Vertebrata. clades, six other lineages containing 3–4 species are Most species of the Melanothamnus clade are completely highly supported (BP/PP > 90/0.95); among these are erect or have a very short prostrate system. However, the Carradoriella clade including Polysiphonia vir- some taxa are decumbent (e.g. Polysiphonia blandii, P. gata, the type species of Caradoriella, and the simplex), forming extensive prostrate systems with rhi- Streblocladia clade, which includes the type species zoids in the basal parts of the erect axes. Members of the S. glomerulata. Our phylogenetic tree also resolved Polysiphonieae are typically smaller than 10 cm. As an five individual species as sisters to the other clades exception, M. afaqhusainii can exceed 1 m in length. with low support. The Melanothamnus clade receives support of 100/ Rhizoids: The connection between the rhizoids and the 1.00 (Fig. 1). In addition to F. unilateralis, N. flavi- pericentral cells from which they originate is a uniform marina and M. somalensis, this clade includes 27 character within each clade, so far as it can be observed other species currently assigned to Neosiphonia and (Table 1). Rhizoids are in open connection with pericen- Polysiphonia. The 95 bp sequence obtained from the tral cells in Polysiphonia sensu stricto (Fig. 19), while they type material of F. unilateralis analysed separately are cut off from pericentral cells in the other clades showed that this sequence was positioned unequivo- (Figs 20–24). Observations on rhizoids cannot be made cally within the Melanothamnus clade, but sequence in mature specimens of some species, such as Vertebrata ambiguities due to the quality of the DNA made it lanosa which is an obligate hemi-parasite that lacks rhi- impossible to determine its precise position. zoids, and Fernandosiphonia unilateralis, Streblocladia The phylogenetic relationships among species glomerulata and Melanothamnus somalensis,whichall within the Melanothamnus clade are generally poorly have compact basal discs without individual rhizoids. resolved, with a few exceptions. Although the lineage formed by ‘Polysiphonia’ schneideri, ‘P.’ amplacapilli, Pericentral cells and cortication: The number of pericen- ‘P.’ pentamera and ‘P.’ morroides is very weakly posi- tral cells and the presence of cortication are variable in tioned as sister to the Melanothamnus clade in the most of the clades (Table 1). All species in the Vertebrata rbcL tree, in 18S analyses this position is robustly clade have six or more pericentral cells, while members supported (see below). of the Polysiphonia sensu stricto clades have four peri- The RAxML 18S tree (Fig. 2) has a similar topol- central cells, with the exception of Bryocladia cuspidata ogy to the rbcL phylogeny, with three strongly sup- (6–8 pericentrals). Cortication is uniformly absent in the ported major clades: Polysiphonia sensu stricto 1 and Polysiphonia sensu stricto and ‘P.’ schneideri clades. 2 and a third clade with all the other taxa. Cortication is variable within the other clades, absent or Polysiphonia sensu stricto clades 1 and 2 are placed slight in small species of Fernandosiphonia but very robustly together (99/1.00). Within the third clade, heavy in Melanothamnus, and absent or slight in most the Vertebrata clade receives full support, while sup- species of Vertebrata with the exception of port is lower for Melanothamnus (82/1.00). The sister Boergeseniella, in which cortication is elaborate. relationship between the Melanothamnus and ‘P.’ schneideri clades is strongly supported in the 18S Plastid arrangement: The arrangement of plastids in phylogeny. In addition, the Carradoriella and the cells is a synapomorphy for the Melanothamnus Streblocladia clades are highly supported. clade. The species in this clade have the plastids lying The time-calibrated phylogenies (Figs S1, S2) esti- exclusively on radial walls of pericentral cells so the mated the divergence in Vertebrata to be more ancient outer walls appear transparent (Table 1, Figs 8, 13, – than in Melanothamnus (90.7–138.66 vs 75.7–95.78 34 39). This particular arrangement of the plastids Ma). Furthermore, the radiation of major lineages in can be easily observed under the microscope as the Vertebrata and Melanothamnus was gradual and took cells show a dark flank when observed in detail place over periods of c. 20 and 12 Ma, respectively. (Figs 14, 35, 37), as well as a transparent halo when the pericentral cells are observed in a suitable posi- tion (Fig. 38). All the other taxa of the Polysiphonieae Morphological observations have plastids against all the cell walls including the – An overview of the distribution of selected morpho- outer wall (Table 1, Figs 25 33). The revision of the logical characters within clades of the Polysiphonieae type materials listed in Table S2, currently assigned to is shown in Table 1. Neosiphonia, allowed us to verify the plastid character in the species Polysiphonia concinna, P. eastwoodiae, Habit: There is considerable variation within and among P. gorgoniae, P. harlandii and P. johnstonii. clades (Table 1), with the exception of the Carradoriella Conversely, the species Lophosiphonia mexicana, P. 14 P. DÍAZ-TAPIA ET AL.

Fig. 2. Phylogenetic tree estimated with ML analysis of 18S sequences. Values at nodes indicate bootstrap support/posterior probability (only shown if > 60%/0.6 PP). Branches marked with an asterisk received 100%/1.00 PP support. Species names printed in bold correspond to type species of genera. EUROPEAN JOURNAL OF PHYCOLOGY 15

Figs 3–8. Melanothamnus somalensis, the type species of Melanothamnus. Fig. 3. Herbarium specimen MICH 662774. Fig. 4. Apical part of a specimen with alternately arranged branches. Figs 5–6. Apices of branches with (Fig. 5) or without (Fig. 6) abundant trichoblasts. Fig. 7. Apex of a lateral branch with trichoblasts. Fig. 8. Surface view of cells with the plastids lying exclusively on radial walls while the outer walls appear transparent (arrows). Scale bars: Fig. 3, 6 cm; Fig. 4, 1 mm; Figs 5 and 6, 350 µm; Fig. 7, 200 µm; Fig. 8, 100 µm.

beaudettei, P. confusa, P. poko, P. profunda and P. character is variable within clades. In the rubrorhiza have the plastids scattered within the cells, Melanothamnus clade this character is absent, and including against the outer wall cells. branches are never axillary.

Branch/trichoblast arrangement: Whether trichoblasts Trichoblast nuclei:Theproximalcellsoftrichoblasts and/or branches are formed on every segment or are are multinucleate in the Vertebrata clade (Table 1, separated by naked segments is variable in Figs 41–43), with up to 8 or more nuclei in the three clades, Melanothamnus, Carradoriella and basal cell and decreasing in number towards the Vertebrata. Most species of the Melanothamnus apices, which can be uninucleate. The nuclei are clade have branches or trichoblasts on every segment, uniformly distributed inside the cells, each appear- which is a key feature of Neosiphonia. However, there ing to have a domain within the cell. Conversely, are exceptions in this clade such as Neosiphonia col- all the cells of trichoblasts are uninucleate in other labens, Polysiphonia nuda and P. pseudovillum from clades of the Polysiphonieae (Figs 40, 44–46). The Panama, in which there are interspersed naked seg- only known exception is Leptosiphonia schousboei, ments. Conversely, the formation of trichoblasts/ which sometimes has two nuclei in the trichoblast branches with naked segments between them is a cells. uniform character in the Streblocladia, ‘Polysiphonia’ schneideri and Polysiphonia sensu stricto clades Branching pattern: Despite the great significance pre- (Table 1). Within all clades except Melanothamnus, viously placed on dorsiventral vs radial branching in the branches may form in a position axillary to tricho- Rhodomelaceae, this character varies within all our blasts, but although constant at the species level, this clades. A primary dorsiventral branching pattern 16 P. DÍAZ-TAPIA ET AL.

Figs 9–18. Fernandosiphonia unilateralis type material, the type species of Fernandosiphonia. Fig. 9. Herbarium specimen. Figs 10–11. Branches unilaterally arranged. Fig. 12. Axis with scar cells of trichoblasts (arrows). Figs 13–14. Surface view of pericentral cells with plastids lying only on the radial walls, so that the outer walls appear transparent (Fig. 13, arrows) and cells have a dark flank (Fig. 14). Fig. 15. Young spermatangial branch formed on the first dichotomy of a trichoblast, the other branch remaining vegetative (arrow). Fig. 16. Procarp (su = supporting cell; cp = carpogonium). Fig. 17. Cystocarp. Fig. 18. Tetrasporangia arranged in short spiral series. Scale bars: Fig. 9, 3 cm; Fig. 10, 2 mm; Fig. 11, 450 µm; Figs 12, 14, 17 and 18, 100 µm; Figs 13 and 15, 40 µm; Fig. 16, 20 µm. characterizes some species of the clades dorsiventral in S. glomerulata, the type species, to spiral Melanothamnus (F. unilateralis and N. collabens; or pseudodichotomous in Polysiphonia muelleriana and Figs 10, 11), Streblocladia (S. glomerulata)and Polysiphonia sp. Likewise, the dorsiventral Neosiphonia Vertebrata (Ctenosiphonia hypnoides)(Table 1). collabens is related to species with spiral or pseudodi- However, this characteristic is not significant in deli- chotomous branching patterns. neating these three genera, as our phylogenetic tree reveals that each of these four species is closely related Spermatangial branches: Whether spermatangial to others that have spirally or pseudodichotomously branches replace trichoblasts completely or replace arranged branches. For example, the branching pattern only one branch of a dichotomously branched tricho- of members of the Streblocladia clade varies from blast is a constant character in all clades (Table 1). In EUROPEAN JOURNAL OF PHYCOLOGY 17

within clades, except for Carradoriella in which they are present (Figs 47–52, Table 1).

Carpogonial branches:TheMelanothamnus clade is characterized by having 3-celled carpogonial branches (Table 1). In our study, we observed this character in N. harveyi, N. collabens, P. blandi and P. forfex (Figs 57, 58). By contrast, the other Polysiphonieae uniformly have 4-celled carpogonial branches like the majority of the Rhodomelaceae (Table 1, Figs 53–56).

Cystocarps: The outline morphology of cystocarps var- ies from globose to ovoid in all the clades analysed here (Table 1, Figs 60–64). Urceolate cystocarps are exclusive to the Polysiphonia sensu stricto clades (Table 1, Fig. 59). Cells around the ostiole are conspicuously larger than (more than twice the size of) the cells of the pericarp immediately below in most species of the Melanothamnus clade (Fig. 70). However, Neosiphonia harveyi is an exception, and the cells of the ostiole in this species are only slightly larger than the other cells of the pericarp. This character is also seen in Streblocladia glomerulata (Fig. 68). Conversely, the cells of the ostiole in the other four clades are uniformly similar to the cells below (Figs 65–67, 69).

Tetrasporangia: The formation of tetrasporangia in straight or spiral rows is variable in all clades (Table 1, – Figs 71 76). It must be noted that very long straight series of tetrasporangia are typically observed only in members of the Polysiphonia sensu stricto clade (Fig. 71). However, straight series can also form in other clades, for example in Neosiphonia collabens and Polysiphonia nuda within the Melanothamnus clade, whereas tetrasporangia in Fernandosiphonia unilatera- lis form short and markedly spiral series (Fig. 18). The third tetrasporangial cover cell is exclusive to the Polysiphonia sensu stricto clade, but this character has not been examined in all the species. Figs 19–24. Rhizoid anatomy in the Polysiphonieae. In open connection with pericentral cells in (Fig. 19, Polysiphonia sensu stricto clade 1). Cut off from pericentral cells in P. foetidissima (Fig. 20, Vertebrata Discussion clade), P. denudata (Fig. 21, Carradoriella clade), Polysiphonia sp. (Fig. 22, Streblocladia clade), P. schneideri Phylogenetic analysis ‘ ’ (Fig. 23, P. schneideri clade) and P. incompta (Fig. 24, Amongst the Polysiphonieae studied here, the early- Melanothamnus clade). Scale bars: Figs 19–23, 100 µm; Fig. 24, 500 µm. branching clade/clades Polysiphonia sensu stricto 1 and 2 were resolved as separate lineages in rbcL analyses (Fig. 1) but together formed a robust clade the two Polysiphonia sensu stricto and Streblocladia in 18S analyses (Fig. 2). The marked discordance clades, spermatangial branches almost uniformly between rbcL and 18S trees regarding the mono- replace trichoblasts (Fig. 47). In the other clades, phyly/paraphyly of the Polysiphonia sensu stricto they are formed on the first dichotomy of modified lineages requires additional research for a more accu- trichoblasts (Figs 15, 49–52), with the exception of rate assessment of relationships and character evolu- Vertebrata (Fig. 48)asV. lanosa has no trichoblasts tion. Because the Polysiphonia sensu stricto lineages in male thalli – they can be observed only occasion- occur near the base of the tree, it is possible that the ally in females. The presence or absence of apical outgroups (which are relatively distant taxa compared sterile cells on spermatangial branches is variable with the ingroup) could have attached to the ingroup 18 P. DÍAZ-TAPIA ET AL.

Figs 25–39. Plastid arrangement in the Polysiphonieae. Scattered against all cell walls of the pericentral cells in Polysiphonia stricta (Figs 25–26, Polysiphonia sensu stricto clade 1), Vertebrata lanosa (Figs 27–28, Vertebrata clade), P. virgata (Figs 29–30, Carradoriella clade), Polysiphonia sp. (Fig. 31, Streblocladia clade) and P. schneideri (Figs 32–33, ‘P.’ schneideri clade). Lying exclusively on the radial walls of the pericentral cells in species of the Melanothamnus clade: Neosiphonia collabens (Figs 34–35), N. harveyi (Figs 36–38) and P. forfex (Fig. 39). Scale bars: Figs 25, 27, 29, 38 and 39, 500 µm; Figs 26, 28 and 30, 800 µm; Figs 31, 32, 34, 35 and 37, 100 µm; Fig. 33, 300 µm; Fig. 36, 50 µm. in the wrong position in one of the analyses (Shavit sensu stricto by the rhizoid anatomy (cut off from et al., 2007). Future work should focus on including a pericentral cells). Both clades are identified by dis- wider range of taxa from across the Rhodomelaceae tinct morphological synapomorphies. The Vertebrata as well as using larger, multi-gene datasets to infer the clade is characterized by the multinucleate cells of correct branching order of the two Polysiphonia sensu trichoblasts; the other key feature of Vertebrata, that stricto lineages. all species have six or more pericentral cells, is shared The Vertebrata and Melanothamnus clades were with members of some other clades. The resolved as large, speciose clades with strong support Melanothamnus clade is unequivocally distinguished using rbcL. The 18S phylogeny also resolves the from other Polysiphonieae by two synapomorphic Vertebrata clade with robust support, while characteristics: the plastid arrangement and the 3- Melanothamnus is moderately well supported. Both celled carpogonial branches. Furthermore, branch clades are clearly distinguished from Polysiphonia origin is independent from trichoblasts in all the EUROPEAN JOURNAL OF PHYCOLOGY 19

Figs 40–46. Trichoblast nuclei (arrows) in the Polysiphonieae. Uninucleate trichoblast cells in Polysiphonia scopulorum (Fig. 40, Polysiphonia sensu stricto clade 1), P. denudata (Fig. 44, Carradoriella clade), P. schneideri (Fig. 45, ‘P.’ schneideri clade) and P. blandii (Fig. 46, Melanothamnus clade). Multinucleate trichoblast cells in species of the Vertebrata clade: P. nigra (Fig. 41), Boergeseniella fruticulosa (Fig. 42) and P. foetidissima (Fig. 43). Scale bars: Figs 40–43, 60 µm, Fig. 44, 30 µm; Fig. 45, 20 µm; Fig. 46, 100 µm. species of this clade, and the majority of species have their phylogenetic relationships and establish a nat- enlarged ostiolar cells. ural classification – it would be premature to spec- In addition to the above-mentioned clades, the ulate on the outcomes of these investigations at rbcL phylogeny resolved six small (3–4 species) but present. One of the major shortcomings in highly supported clades, as well as indicating five Polysiphonieae sequence databases is the uneven species that are uncertainly positioned. The generic geographical sampling, as the majority of sequenced assignment of these lineages requires further taxon taxa come from Atlantic and central to and gene sampling in order to better understand north-estern America. The generation of molecular 20 P. DÍAZ-TAPIA ET AL.

data from additional regions could contribute to acquiring a more realistic perspective of the magni- tude of unplaced lineages and to delineating their corresponding genera. Also, the resolution of the commonly employed molecular markers in the Polysiphonieae is not sufficient to resolve the phy- logenetic relationships among numerous lineages, which could be improved using larger gene datasets (Díaz-Tapia et al., 2015).

Taxonomic position of Vertebrata Vertebrata lanosa is placed in a strongly supported clade that also includes Brongniartella byssoides, Boergeseniella fruticulosa, Ctenosiphonia hypnoides and Enelittosiphonia stimpsonii, the type species of their corresponding genera, and Lophosiphonia repta- bunda (which is not the type species). All members of this clade have a synapomorphic characteristic that was previously overlooked in relation to systematics (but see Maggs & Hommersand, 1993): multinucleate trichoblast cells. We conclude from molecular and morphological evidence that members of this clade represent a single genus. Vertebrata is the oldest name among those available for this clade, as noted before (Choi et al., 2001), and the new combinations proposed in Table 2 are required. Furthermore, the Vertebrata binomials previously established by

Kuntze (1891) should be reinstated for the other 13 species included in this clade (Table 3). Interestingly, Brongniartella is not monophyletic despite its distinctive persistent and pigmented tri- choblasts that led to its classification in the tribes Lophothalieae (Falkenberg, 1901; Womersley, 2003) or Brongniartelleae Parsons (Parsons, 1975; Maggs & Hommersand, 1993). Although trichoblasts are typically considered unpigmented in the Polysiphonieae, they are com- monly pigmented when young before they enlarge and become colourless (Delivopoulos, 2002). The two currently recognized species of Brongniartella, B. byssoides and B. australis, were separated within the Vertebrata clade, respectively placed with V. lanosa and Polysiphonia nigra. Ctenosiphonia is a monotypic genus segregated from Polysiphonia due to its very peculiar morpholo- Figs 47–52. Spermatangial branches in the Polysiphonieae. gical characteristics, including a dorsiventral thallus Replacing trichoblasts and with sterile apical filaments in Polysiphonia stricta (Fig. 47, Polysiphonia sensu stricto clade and two tetrasporangia per segment (Falkenberg, 1). Replacing trichoblasts and lacking sterile apical cells in 1901; Díaz-Tapia & Bárbara, 2013). This genus, Vertebrata lanosa (Fig. 48, Vertebrata clade). On a branch of together with Lophosiphonia Falkenberg (1897), is a trichoblast and with sterile apical cells in P. fucoides (Fig. 49, currently positioned within the ‘Lophosiphonia Vertebrata clade), P. denudata (Fig. 50, Carradoriella clade), P. group’ (Falkenberg, 1901). Boergeseniella and schneideri (Fig. 51, ‘P.’ schneideri clade) and Neosiphonia har- veyi (Fig. 52, Melanothamnus clade). Scale bars: 100 µm. Enelittosiphonia were distinguished from other Arrows show the apical sterile cells and arrowheads the sterile Polysiphonieae by their particular branching patterns branch of fertile trichoblasts. (Segi, 1949; Kylin, 1956), but our molecular evidence EUROPEAN JOURNAL OF PHYCOLOGY 21

Figs 53–58. Carpogonial branches in the Polysiphonieae. Four-celled in Polysiphonia stricta (Fig. 53, Polysiphonia sensu stricto clade 1), P. nigra (Fig. 54, Vertebrata clade), P. denudata (Fig. 55, Carradoriella clade) and P. schneideri (Fig. 56, ‘P.’ schneideri clade). Three-celled in species of the Melanothamnus clade: Neosiphonia harveyi (Fig. 57) and P. blandii (Fig. 58). Su = supporting cell; st = sterile basal cell; 1–4 cells of carpogonial branches. Scale bars: Fig. 53, 30 µm; Figs 54–58, 20 µm.

(Fig. 1) does not support their recognition as inde- important at levels of genus and tribe (Falkenberg, pendent genera. 1901; Kylin, 1956; Hommersand, 1963). The diversity of currently recognized genera The main morphological character delineating the included in this clade reflects the high variability Vertebrata group is that trichoblast cells are multi- among Vertebrata species in trichoblast characteris- nucleate. In the Polysiphonieae and some other tics (pigmented/unpigmented; persistent/deciduous; Ceramiales, the apical cell is uninucleate, whereas spirally/dorsiventrally arranged) and branching pat- the cells cut off from it undergo nuclear divisions terns (spiral/dorsiventral; presence or absence of during elongation, becoming multinucleate, with the alternating branches of determinate and indetermi- number of nuclei being proportional to the volume of nate growth), which classical authors considered the cell (Goff & Coleman, 1986; McIvor et al., 2002). 22 P. DÍAZ-TAPIA ET AL.

Figs 65–70. Cells surrounding the ostiole in the Polysiphonieae. Similar or slightly larger than the cells of the pericarp inmediately below in Polysiphonia stricta (Fig. 65, Polysiphonia sensu stricto clade 1), Vertebrata lanosa (Fig. 66, Vertebrata clade), P. denudata (Fig. 67, Carradoriella clade), and P. schneideri (Fig. 69, ‘P.’ schnei- deri clade). They are much larger in Streblocladia glomer- ulata (Fig. 68, Streblocladia clade) and Neosiphonia collabens (Fig. 70, Melanothamnus clade). Scale bars: Figs 65–68 and 70, 100 µm; Fig. 69, 60 µm.

the vegetative hairs of the red algae or trichoblasts of the Rhodomelaceae including desiccation resis- tance, nutrient uptake, metabolite secretion, shading, trapping of spermatia, mucilage stabilization and monitoring of phosphorus status (Delivopoulos, Figs 59–64. Cystocarps in the Polysiphonieae. Urceolate in 2002, and references therein). Physically, trichoblasts Polysiphonia stricta (Fig. 59, Polysiphonia sensu stricto clade can form a dense network around the apices that 1). Ovoid in Vertebrata lanosa (Fig. 60, Vertebrata clade), P. could potentially restrict access to the cells by small denudata (Fig. 61, Carradoriella clade), Streblocladia glo- grazers, such as amphipods and copepods. merulata (Fig. 62, Streblocladia clade). Globose in Polysiphonia schneideri (Fig. 63, ‘P.’ schneideri clade) and Neosiphonia collabens (Fig. 64, Melanothamnus clade). Scale Taxonomic position of Neosiphonia, bars: Figs 59–62 and 64, 200 µm; Fig. 63, 100 µm. Fernandosiphonia and Melanothamnus The presence of the type species of the genus The trichoblasts of the Rhodomelaceae are usually Melanothamnus (M. somalensis) in a strongly sup- uninucleate, whereas the polysiphonous parts of the ported clade with the type species of the genus thalli are multinucleate (Coomans & Hommersand, Neosiphonia (N. flavimarina) and Fernandosiphonia 1990; Garbary & Clarke, 2001; Delivopoulos, 2002). A (F. unilateralis) indicates that Neosiphonia, plausible advantage of having multinucleate tricho- Fernandosiphonia and Melanothamnus are not dis- blasts in Vertebrata is that their cells can reach larger tinct monophyletic genera. Neosiphonia is a later sizes. In fact, trichoblasts in this genus are sometimes heterotypic synonym of Fernandosiphonia. However, extremely well developed, exceeding 10 mm in length the name Melanothamnus is older than both in species such as Vertebrata (Lophosiphonia) repta- Fernandosiphonia and Neosiphonia, and the new bunda and V.(Ctenosiphonia) hypnoides. In the red combinations proposed in Tables 4 and 5 are algae, cell streaming is slow compared with other required. These new combinations include 31 species algae (Pueschel, 1990), and multinuclearity of large that were previously assigned to Neosiphonia; two cells may facilitate the regulation of cellular activities. species known to be closely related to Neosiphonia Several potential functions have been attributed to but that had been retained in Polysiphonia because EUROPEAN JOURNAL OF PHYCOLOGY 23

their morphology conflicted with Kim & Lee (1999); six species for which molecular data are presented here for the first time; and three species that are transferred to Melanothamnus on the basis of their morphology. On the other hand, 10 species that are currently placed in Neosiphonia should be replaced in Polysiphonia for formal purposes pending clarifica- tion of their phylogenetic affinities and generic assignment. Polysiphonia beaudettei, P. confusa, P. echinata, P. elongella, P. poko, P. rubrorhiza and P. profunda were assigned to Neosiphonia based on morphological characteristics (Kim & Lee, 1999; Abbott et al., 2002; Kim & Abbott, 2006; Mamoozadeh & Freshwater, 2011; Norris, 2014). However, they lack the plastid character, and further- more molecular data for P. echinata and P. elongella show that they do not belong to the Melanothamnus clade (Fig. 1). Likewise, Polysiphonia sertularioides was transferred to Neosiphonia based on the mor- phology of Korean material attributed to this species (Nam & Kang, 2012). However, its type locality is in the Mediterranean, and Atlantic sequences for this species are not in the Fernandosiphonia clade (Fig. 1; Mamoozadeh & Freshwater, 2012). Polysiphonia paniculata was transferred to Neosiphonia (Norris, 2014), but again it is not in the Melanothamnus clade (Figs 1, 2). Finally, our study of

the type material of Lophosiphonia mexicana, also transferred to Neosiphonia (Norris, 2014; Table S2), indicates that this species is probably not a member of the Polysiphonieae. As noted by Norris (2014), further studies are needed to clarify the generic pla- cement of this unusual species, and meanwhile we propose to leave it in Lophosiphonia until more infor- mation is available. Specimens of Melanothamnus collected in Oman and housed in MICH were initially assigned to M. somalensis (Wynne & Banaimoon, 1990), before the description of M. afaqhusainii from Pakistan (Shameel, 1999). Revision of the Omani materials leads us to conclude that both M. somalensis and M. afaqhusainii are represented in Oman, and their mor- phology agrees with the criteria proposed by Shameel (1999) and Afaq-Husain & Shameel (2000) for distin- guishing them. Their rbcL sequences diverged by 1.4% (18 bp). The Melanothamnus clade is morphologically dis- tinguished from other members of the tribe Figs 71–76. Tetrasporangia in the Polysiphonieae. Forming Polysiphonieae by an unequivocal synapomorphic long straight series in Polysiphonia stricta (Fig. 71, character: plastids lie exclusively on the radial walls Polysiphonia sensu stricto clade 1). Forming spiral series in of the pericentral cells and are absent from outer Vetebrata lanosa (Fig. 72, Vertebrata clade), Polysiphonia sp. (Fig. 74, Streblocladia clade) and Neosiphonia harveyi (Fig. 76, walls. The plastid character was previously noted by ‘ Melanothamnus clade). Forming short straight series in P. Hollenberg (1961, 1968a), who described hyaline cell denudata (Fig. 73, Carradoriella clade), and P. schneideri walls’ for several species (e.g. P. pseudovillum, P. (Fig. 75, P. schneideri clade). Scale bars: Figs 71, 74 and 76, bajacali), and by Maggs & Hommersand (1993). 200 µm; Figs 72, 73 and 75, 400 µm. However, its significance at higher taxonomic levels 24 P. DÍAZ-TAPIA ET AL. has not previously been highlighted. We observed The key morphological feature of Melanothamnus this character in a total of 35 species, and we con- is the restriction of plastids to the radial walls of the clude that it is uniform in the Melanothamnus clade. pericentral cells and their absence from the outer Conversely, other Polysiphonieae and most of the walls. Algae demonstrate a notable decline in photo- Rhodomelaceae have plastids distributed within the synthesis at higher light levels possibly due to damage cytoplasm, some lying against outer cell walls. In the to the photosynthetic apparatus caused by excessive family, the only other exception is some species of light delivery to photosystem II (Lüning, 1990; Hurd Herposiphonia in which the plastids form transverse et al., 2014). Many green and brown algal plastids bands (Hollenberg, 1968c; Womersley, 2003; Díaz- have phototropic reactions to blue and UV light in Tapia & Bárbara, 2013). order to protect them from irradiation damage Carpogonial branches are typically 4-celled through- (Lüning, 1990). Plastid movement, however, has out the family Rhodomelaceae. Three-celled carpogo- never been demonstrated for the vast majority of nial branches were described for the first time in red algae (Pueschel, 1990), and it appears that red Polysiphonia platycarpa (Iyengar & Balakrishnan, algae have evolved other types of protection against 1950), and later this was one of the features proposed UV damage. Red algae including Polysiphonia species to delineate the genus Neosiphonia (Kim&Lee,1999). have high concentrations of various mycosporine-like Three-celled carpogonial branches have been reported amino acids (MAAs) that respond rapidly to envir- in 17 species (four of them in the ‘japonica-complex’), onmental change and act as defences against the all of which are placed here in Melanothamnus. photooxidative effects of sunlight (Karsten et al., Alternative interpretations of the carpogonial branch 1998; Navarro et al., 2014). The movement of the configuration were found in the literature for F. uni- plastids onto the radial walls of the periaxial cells, in lateralis, as Levring (1941) described and illustrated a 4- combination with MAAs, may have given the celled structure, while Morrill (1976, plate 37, figs E and Melanothamnus ancestor a selective advantage over H) illustrated 3-celled carpogonial branches in the same other Polysiphonieae, allowing it to exploit new eco- species, also from the type locality. This character can logical niches. The prevalence of Melanothamnus be easily misinterpreted if the carpogonial branches are species in habitats with exposure to high light levels, not observed at the right developmental stage. In our such as in Hawaii or turfs on coral reefs (Price & study of the type material, a single procarp was observed Scott, 1992; Kim & Abbott, 2006), supports this in a permanent slide (Fig. 16), most probably the same speculation. one illustrated by Levring. It is unclear how this procarp should be interpreted because it is too mature, and so it Biogeography of Vertebrata and Melanothamnus is difficult to determine with certainty which cell corre- sponds to the sterile basal cell and which to the basal cell The genus Vertebrata is distributed worldwide, and of the carpogonial branch. From the evidence of the representatives have been described from all regions presence of both the plastid character and 3-celled where there has been a detailed study of the tribe carpogonial branches, Kintarosiphonia fibrillosa Uwai Polysiphonieae. The majority of our sequences are &Masuda(1999), based on Pterosiphonia fibrillosa,and from Europe, but our systematic review and unpub- Polysiphonia platycarpa are also here transferred to lished data suggest that this genus is widespread. Melanothamnus (Table 1). BEAST and PhyloBayes calibrations indicate radiation The other morphological characters proposed by of the major lineages of the Vertebrata clade over a 20 Kim&Lee(1999) to delineate the genus MA period starting about 140 or 90 Ma (estimates from Neosiphonia vary among closely related species, different methods; see Figs S1, S2). Further conclusions except for the rhizoid anatomy. Rhizoids are cut as to its origins and centres of diversity would be pre- off from the pericentral cells in all Polysiphonieae mature, pending more comprehensive sampling. except for Polysiphonia sensu stricto in which they In contrast, Melanothamnus is predominantly Indo- areinopenconnectionwith the pericentral cells. Pacific (Fig. 77). Although few molecular data are avail- After the establishment of Neosiphonia,numerous able from Indian coasts, some species occur in South species were transferred to the new genus based on Africa (M. incompta), Oman (M. somalensis and M. morphology, but commonlyoverlookingthenum- afaqhusainii), India (M. platycarpa) and Thailand (M. ber of cells in carpogonial branches. Excluding this thailandica). Among the regions for which there is a trait, several species have all five characteristics pro- comprehensive study of the Polysiphonieae, the diver- posed by Kim & Lee (1999) to delineate Neosiphonia sity of Melanothamnus is particularly high in Korea, but nevertheless are not in the Melanothamnus clade Japan and Hawaii (14, 11 and 14 species, respectively). (e.g. P. brodiei, P. echinata, P. elongella), while sev- This genus is also well represented on North American eral species are clearly in the clade (e.g. M. collabens, Atlantic coasts (4–5 species), but it is almost completely M.nuda,M.pseudovillum) but lack this combina- absent from Atlantic and Mediterranean Europe, where tion of traits. only two species have been reported, which are both EUROPEAN JOURNAL OF PHYCOLOGY 25

Fig. 77. World map representing the proportion of Melanothamnus (black) and other Polysiphonieae (grey) species in selected regions where the Polysiphonieae were studied in detail. Data were obtained from the following references after updating the species names: Alaska: Lindstrom (http://www.seaweedsofalaska.com); Brazil (Espírito Santo-São Paulo): Guimâraes et al.(2004); California: Abbott & Hollenberg (1976); Florida: Schneider & Searles (1991); Hawaii: Abbott (1999); Japan: Yoshida (1998); Korea: Nam & Kang (2012); Panama: Mamoozadeh & Freshwater (2012); South Africa: Stegenga et al. (1997); southern France: Lauret (1967, 1970) Spain (Galicia): Bárbara et al.(2005); British Isles: Maggs & Hommersand (1993). most probably examples of old human-mediated intro- the Tethys Seaway, between 60 and 20 million years ago, ductions. Melanothamnus harveyi is native to south- particularly as the sister ‘P.’ schneideri clade is also eastern Asia and has been introduced by multiple events primarily Pacific in distribution. In our rbcL phylogeny onto northern Atlantic coasts (McIvor et al., 2001; (Fig. 1), the ‘P.’ schneideri clade includes two Korean Savoie & Saunders, 2016). Similarly, M. collabens is species and two species distributed in the Pacific and likely to be an old introduction into the Atlantic, North America with one of them introduced in Europe where it extends from the Bay of Biscay to Cape (Díaz-Tapia et al., 2013a). Furthermore, our surveys in Verde, including the western Mediterranean (Díaz- Australia revealed five other Indo-Pacific species Tapia & Bárbara, 2013). The finding of M. collabens in belonging to this clade (unpublished data). However, California (as P. johnstonii,seeTable 4) supports this BEAST and PhyloBayes calibrations indicate radiation hypothesis, but although Polysiphonia johnstonii was of the major lineages of the Melanothamnus clade over a first collected from the Gulf of California in 1921 12 Ma period starting about 95 or 75 Ma (the two (Setchell & Gardner, 1924), California was probably methods providing different estimates), with divergence not the original source of the introduction. An investi- from the ‘P.’ schneideri clade 140 or 95 Ma (Figs S1, S2). gation of Polysiphonia species from the Northern Gulf The distribution resembles a Tethyan one that origi- of California (Hollenberg & Norris, 1977)suggested nated during the Cretaceous 125– 75 Ma (Lüning, that since its initial collection and description, P. john- 1990;Hommersand,2007) when the Tethys Ocean stonii has extended its range along the Pacific coast of formed a tropical girdle around the earth. Unlike typical North America, fulfilling one of the criteria for an Tethyan distributions, in addition to its wide occurrence invasive species (Chapman & Carlton, 1991;Ribera& throughout the tropics, Melanothamnus occupies more Boudouresque, 1995). How far this species has spread temperate regions in the North Pacific (e.g. Japan, along the Pacific coast of America and along North Korea) and the South Pacific/Oceania (e.g. South Atlantic coastlines remains to be determined, and Australia, New Zealand). The question of whether further sampling is needed to establish its origin. Melanothamnus failed to colonize the north-eastern The absence of naturally occurring Melanothamnus Atlantic as it opened up during the Cretaceous, or species in the Mediterranean and north-eastern Atlantic whether north-eastern Atlantic lineages evolved but might suggest that Melanothamnus is of recent origin, became extinct, perhaps during Pleistocene glaciations, having evolved in the Pacific Ocean after the closure of cannot be answered at present. 26 P. DÍAZ-TAPIA ET AL.

Acknowledgements collection and study of Melanothamnus somalensis and M. afaqhusianii; editing manuscript; C. Maggs: original WearepleasedtorecognizeherethatProfessorMaxH. concept, drafting and editing manuscript. Hommersand (UNC Chapel Hill) independently discovered the key morphological character of the position of plastids in the Melanothamnus clade. We thank Joana Costa for providing ORCID us samples of Polysiphonia spp. from South Africa, and Dr Heroen Verbruggen http://orcid.org/0000-0002-6305- Erasmo Macaya for supplying material from his recent collec- 4749 tions of Fernandosiphonia unilateralis from the type locality. Christine A. Maggs http://orcid.org/0000-0003-0495- We are grateful to Dr Cynthia Trowbridge for providing sam- 7064 ples from Japan, and Dr Show-Mei Lin for collecting samples in Taiwan. Professor Gary Saunders is warmly thanked for generously sharing sequences for Melanothamnus afaqhusainii fromOman.Weacknowledgethecuratorsoftheherbaria BM(NH),TFC,USandTCD;DrJoeZuccarelloprovided References host support for Lynne McIvor during her Marie Curie fellow- Abbott, I.A. (1999). Marine Red Algae of the Hawaiian ship at Leiden Herbarium. We thank two anonymous Islands. Bishop Museum Press, Honolulu. reviewers for their valuable comments on the manuscript. Abbott, I.A. & Hollenberg, G.J. (1976). Marine Algae of California. Stanford University Press, Stanford, CA. Abbott, I.A., Fisher, J. & McDermid, K.J. (2002). New Disclosure statement reported and revised marine algae from the vicinity of Nha Trang, Vietnam. In of Economic No potential conflict of interest was reported by the author(s). Seaweeds with Reference to some Pacific Species (Abbott, I.A. & McDermid, K. J., editors), 291–321. California Sea Grant College, La Jolla. Funding Adams, N.M. (1991). The New Zealand species of Polysiphonia Greville (Rhodophyta). New Zealand PDT acknowledges support by the postdoctoral programmes Journal of Botany, 29: 411–427. Axudas de apoio á etapa inicial de formación posdoutoral do Adams, N.M. (1994). Seaweeds of New Zealand. An Plan I2C (Xunta de Galicia). Funding for the aspects of the Illustrated Guide. Canterbury University Press, work carried out in the Melbourne lab was provided by the Christchurch. Australian Biological Resources Study (RFL213-08) and the Afaq-Husain, S. & Shameel, M. (2000). Further investiga- Australian Research Council (FT110100585). tions on the red alga Melanothamnus afaqhusainii (Ceramiales) from the coast of Pakistan. Pakistan Journal of Botany, 32:15–26. Supplementary information Agardh, C.A. (1824). Systema algarum. Berlinginiana, Lundae. The following supplementary material is accessible via the Agardh, J.G. (1842). Algae maris Mediterranei et Adriatici, Supplementary Content tab on the article’s online page at observationes in diagnosin specierum et dispositionem http://dx.doi.org/10.1080/09670262.2016.1256436 generum. Fortin, Masson et Cie, Paris. Supplementary Table S1. Sample information for the Agardh, J.G. (1863). Species Genera et Ordines Algarum, seu species included in the phylogenetic analysis. descriptiones succinctae specierum, generum et ordinum, Supplementary Table S2. List of herbarium specimens quibus algarum regnum constituitur. Volumen secundum: examined for comparison with Melanothamnus species. algas florideas complectens. C.W.K. Gleerup, Lundae. Supplementary Table S3. Percentage of bases that differ Bailey, J.W. (1848). Continuation of the list of localities of (below) and are identical (above) for the species sequences algae in the United States. American Journal of Science of the Melanothamnus clade included in our rbcL and Arts, series 2, 6:37–42. alignment. Bárbara, I., Cremades, J., Calvo, S., López-Rodríguez, M.C. Supplementary Table S4. Percentage of bases that differ & Dosil, J. (2005). Checklist of the benthic marine and (below) and are identical (above) for the species sequences brackish Galician algae (NW Spain). Anales del Jardín of the Vertebrata clade included in our rbcL alignment. Botánico de Madrid, 62:69–100. Supplementary Fig. S1. Chronogram resulting from the Bárbara, L, Choi, H.-G., Secilla, A., Díaz-Tapia, P. Bayesian relaxed molecular clock analysis performed with Gorostiaga, J.M., Seo, T.-K., Jung, M.Y. & Berecibar, E. BEAST. (2013). Lampisiphonia iberica gen. et sp. nov. Supplementary Fig. S2. Chronogram resulting from the (Ceramiales, Rhodophyta) based on morphology and autocorrelated molecular clock analysis performed with molecular evidence. Phycologia, 52: 137–155. PhyloBayes. Børgesen, F. (1918). The marine algae of the Danish West Indies. Part 3. Rhodophyceae (4). Dansk Botanisk Arkiv, – Author contributions 3: 241 304. Børgesen, F. (1934). Some Indian Rhodophyceae especially P. DÍaz-Tapia: original concept, drafting and editing from the shores of the Presidency of Bombay-IV. manuscript; determination and analysis of molecular data; Bulletin of Miscellaneous Information, Royal Botanic L. McIvor: original concept; determination of molecular Gardens, Kew, 1934:1–30. data; D.W. Freshwater: determination and analysis of Bornet, É. (1892). Les algues de P. K. A. Schousboe, molecular data; editing manuscript; H. Verbruggen: analy- récoltées au Maroc & dans la Méditerranée de 1815 a sis of molecular data; editing manuscript; M.J. Wynne: 1829, et déterminées par M. Édouard Bornet. Mémoires EUROPEAN JOURNAL OF PHYCOLOGY 27

de la Société Nationale des Sciences naturelles et Falkenberg,P.(1901). Die Rhodomelaceen des Golfes von Mathématiques de Cherbourg, 28: 165–376. Neapel und der angrenzenden Meeres-Abschnitte. Fauna Bory de Saint-Vincent, J.B.G.M. (1822). Brongniartelle. und Flora des Golfes von Neapel, Monographie 26. Brongniartella. Bot. crypt. (Céramiaires.). In Berlin. Dictionnaire Classique d’Histoire Naturelle (Audouin, I. Freshwater, D.W. & Rueness, J. (1994). Phylogenetic rela- et al., editors), 516–517. Rey et Gravier; Baudoin frères, tionships of some European Gelidium (Gelidiales, Paris. Rhodophyta) species, based on rbcL nucleotide sequence Bustamante, D.E., Won, B.Y., Ramirez, M.E. & Cho, T.O. analysis. Phycologia, 33: 187–194. (2012). Neosiphonia peruviensis sp. nov. (Rhodomelaceae, Garbary, D.J. & Clarke, B. (2001). Apoptosis in trichoblast Rhodophyta) from the Pacific coast of South America. development in Polysiphonia harveyi (Rhodophyta). Botanica Marina, 55: 359–366. Phycologia, 40: 324–329. Bustamante, D.E., Won, B.Y. & Cho, T.O. (2013a). Goff, L.J. & Coleman, A.W. (1986). A novel pattern of Neosiphonia ramirezii sp. nov. (Rhodomelaceae, apical cell polyploidy, sequential polyploidy reduction Rhodophyta) from Peru. Algae, 28:73–82. and intercellular nuclear transfer in the red alga Bustamante, D.E., Won, B.Y. & Cho, T.O. (2013b). Polysiphonia. American Journal of Botany, 73: 1109– Neosiphonia baliana sp. nov. and N. silvae sp. nov. 1130. (Rhodomelaceae, Rhodophyta) from Bali, Indonesia. Goff, L.J. & Moon, D.A. (1993). PCR amplification of Botanica Marina, 56: 515–524. nuclear and plastid genes from algal herbarium speci- Chapman, J.W. & Carlton, J.T. (1991). A test of criteria for mens and algal spores. Journal of Phycology, 29: 381– introduced species: the global invasion by the isopod 384. Synidotea laevidorsalis (Miers, 1881). Journal of Gray, S.F. (1821). A Natural Arrangement of British Plants. Crustacean Biology, 11: 386–400. . . . Baldwin, Cradock & Joy, Paternoster-Row, London. Choi, H.G., Kim, M.S., Guiry, M.D. & Saunders, G.W. (2001). Greville, R.K. (1824). Flora Edinensis . . .. William Phylogenetic relationships of Polysiphonia (Rhodo- Blackwood, Edinburgh, & T. Cadell, London. melaceae, Rhodophyta) and its relatives based on anatomi- Grunow, A. (1874). Algen der Fidschi-, Tonga- und cal and nuclear small-subunit rDNA sequence data. Samoa-Inseln, gesammelt von Dr. E. Graeffe. Journal Canadian Journal of Botany, 79:1465–1476. des Museum Godeffroy, 3:23–50. Coomans, R.J. & Hommersand, M.H. (1990). Vegetative Guimarâes, S.M.P.D.B., Fujii, M.T., Pupo, D. & Yokoya, N. growth and organization. In Biology of the Red Algae S. (2004). Reavaliaçao das características morfológicas e (Cole, K.M. & Sheath, R.G., editors), 275–304. suas implicaçoes taxonômicas no gênero Polysiphonia Cambridge University Press, Cambridge. sensu lato (Ceramiales, Rhodophyta) do litoral dos esta- Dawes, C.J. & Mathieson, A.C. (2008). The Seaweeds of dos de Sao Paulo e Espírito Santo, Brasil. Revista Florida. University Press of Florida, Gainesville, FL. Brasileira de Botânica, 27: 163–183.

Dawson, E.Y. (1944). The marine algae of the gulf of Guiry, M.D. & Guiry, G.M. (2016). AlgaeBase. World-wide California. The University of Southern California electronic publication, National University of Ireland, Publications, 3: 189–359. Galway. http://www.algaebase.org; searched on 7 Delivopoulos, S. (2002). Ultrastructure of trichoblasts in February 2016. the red alga spectabilis var. spectabilis Hariot, P. (1891). Liste des algues marines rapportés de (Rhodomelaceae, Ceramiales). European Journal of Yokoska (Japon) par M. le Dr Savatier. Mémoires de la Phycology, 37: 329–338. Société nationale des sciences naturelles de Cherbourg, 27: Díaz-Tapia, P. & Bárbara, I. (2013). Seaweeds from sand- 211–230. covered rocks of the Atlantic Iberian Peninsula. Part 1. Harvey, W.H. (1847). Nereis Australis, or Algae of The The Rhodomelaceae (Ceramiales, Rhodophyta). Southern Ocean . . . [Part 1]. Reeve Brothers, London. Cryptogamie, Algologie, 34: 325–422. Harvey, W.H. (1853). Nereis boreali-americana; or, contri- Díaz-Tapia, P., Kim, M.S., Secilla, A., Bárbara, I. & butions towards a history of the marine algae of the Cremades, J. (2013a). Taxonomic reassessment of Atlantic and Pacific coasts of North America. Part II. Polysiphonia foetidissima (Rhodomelaceae, Rhodo- Rhodospermeae. Smithsonian Contributions to phyta) and similar species, including P. schneideri,a Knowledge,5:[i–ii], [1]–258, pls XIII–XXXVI. newly introduced species in Europe. European Journal Harvey, W.H. (1855). Algae. In The botany of the Antarctic of Phycology, 48: 345–362. voyage of H.M. discovery ships Erebus and Terror, in the Díaz-Tapia, P., Bárbara, I. & Berecibar, E. (2013b). years 1839–1843, under the command of Captain Sir Vegetative and reproductive morphology of James Clark Ross. II. Flora Novae-Zelandiae. Part II. Polysiphonia tripinnata (Rhodmelaceae, Rhodophyta): a Flowerless plants (Hooker, J.D., editor), pp. 211–266. new record from the European Atlantic coast. Botanica Reeve, London. Marina, 56: 151–160. Harvey, W.H. (1857 ‘1856’). Algae. In: Account of the Díaz-Tapia, P., Maggs, C.A., West, J.A. & Verbruggen, H. Botanical specimens (Gray, A., editor) Narrative of the (2015). Tackling rapid radiations with chloroplast phy- expedition of an American squadron to the China Seas logenomics in the Rhodomelaceae. European Journal of and Japan, performed in the years 1852, 1853 and 1854, Phycology, 50:54–55. under the command of Commodore M.C. Perry, United Doyle, J.J. & Doyle, J.L. (1987). A rapid DNA isolation States Navy. Volume II – with illustrations. (Anon, edi- procedure for small quantities of fresh leaf tissue. tors), pp. 331–332. Senate of the Thirty-third Congress, Phytochemical Bulletin, 1:11–5. Second Session, Executive Document. House of Drummond, A.J., Ho, S.Y.W., Phillips, M.J. & Rambaut, A. Representatives, Washington. (2006). Relaxed phylogenetics and dating with confi- Harvey, W.H. (1859). Phycologia Australica. Vol. 2. Lovell dence. PLoS Biology, 4: e88. Reeve & Co, London. 28 P. DÍAZ-TAPIA ET AL.

Harvey, W.H. (1860). Characters of new algae, chiefly from organization and analysis of sequence data. Japan and adjacent regions, collected by Charles Wright Bioinformatics, 28:1647–1649. in the North Pacific Exploring Expedition under Captain Kim, M.S. (2003). Taxonomic reassessment of the genus James Rodgers. Proceedings of the American Academy of Polysiphonia (Rhodomelaceae, Rhodophyta): morphol- Arts and Sciences, 4: 327–335. ogy and reproduction of Neosiphonia decumbens and Harvey, W.H. (1862). Phycologia australica.Vol. 4. Lovell N. harlandii. Algae, 18: 107–119. Reeve & Co, London. Kim, M.S. (2005). Taxonomy of a poorly documented alga, Hollenberg, G.J. (1942). An account of the species of Neosiphonia savatieri Rhodomelaceae, Rhodophyta) Polysiphonia on the Pacific coast of North America. I. from Korea. Nova Hedwigia, 81: 163–176. Oligosiphonia. American Journal of Botany, 29: 772–785. Kim, M.S. & Abbott, I.A. (2006). Taxonomic notes on Hollenberg, G.J. (1944). An account of the species of Hawaiian Polysiphonia, with transfer to Neosiphonia Polysiphonia on the Pacific coast of North America. II. (Rhodomelaceae, Rhodophyta). Phycological Research, Polysiphonia. American Journal of Botany, 31: 474–483. 54:32–39. Hollenberg, G. J. (1961). Marine red algae of Pacific Kim, M.S. & Lee, I.K. (1999). Neosiphonia flavimarina gen. Mexico, Part 5: The genus Polysiphonia. Pacific et sp. nov. with a taxonomic reassessment of the genus Naturalist, 2: 345–375. Polysiphonia (Rhodomelaceae, Rhodophyta). Hollenberg, G.J. (1968a). An account of the species of Phycological Research, 47: 271–281. Polysiphonia of the central and western tropical Pacific Kim, B. & Kim, M.S. (2014). Three new species of ocean. II Oligosiphonia. Pacific Science, 22:56–98. Polysiphonia sensu lato (Rhodophyta) based on the mor- Hollenberg, G.J. (1968b). An account of the species of phology and molecular evidence. Algae, 29: 183–195. Polysiphonia of the central and western tropical Pacific Kim, B. & Kim, M.S. (2016). Neosiphonia yongpilii sp. nov. ocean. I. Polysiphonia. Pacific Science, 22: 198–207. (Rhodomelaceae, Rhodophyta), known as Neosiphonia Hollenberg, G. J. (1968c). An account of the species of the red simplex from Korea, with an emphasis on cystocarp alga Herposiphonia occurring in the Central and Western development. Phycologia, 55: 323–332. tropical Pacific Ocean. Pacific Science, 22:536–559. Kim, M.S. & Yang, E.C. (2006). Taxonomy and phylogeny Hollenberg, G.J. & Norris, J.N. (1977). The red alga of Neosiphonia japonica (Rhodomelaceae, Rhodophyta) Polysiphonia (Rhodomelaceae) in the Northern Gulf of based on rbcL and cpeA/B gene sequences. Algae, 21: California. Smithsonian Contributions to the Marine 287–294. Sciences, 1:1–21. Kim, M.S., Lee, I.K. & Boo, S.M. (1994). Morphological Hommersand, M.H. (1963). The morphology and classifi- studies of the red alga Harvey on cation of some Ceramiaceae and Rhodomelaceae. the Korean coast. Korean Journal of Phycology, 9 :185–192. University of California Publications in Botany, 35: Kim, M.-S., Maggs, C.A., McIvor, L. & Guiry, M.D. (2000). 165–366. Reappraisal of the type species of Polysiphonia

Hommersand, M.H. (2007). Global biogeography and rela- (Rhodomelaceae, Rhodophyta). European Journal of tionships of the Australian marine macroalgae. In Algae Phycology, 35:83–92. of Australia (McCarthy, P.M. & Orchard, A.E., editors), Kirby, R.R. & Reid, P.C. (2001). PCR from the CPR offers a 511–542. Australian Biological Resources Study, historical perspective on marine population ecology. Canberra. Journal of the Marine Biological Association of the Hooker, J.D. & Harvey, W.H. (1845). Algae Novae United Kingdom, 81: 539–540. Zelandiae. London Journal of Botany, 4: 521–551. Kraft, G.T. & Abbott, I.A. (2002). The anatomy of Hudson, W. (1762). Flora anglica; exhibens plantas per Neotenophycus ichthyosteus gen. et sp. nov. regnum angliae sponte crescentes, distributas secundum (Rhodomelaceae, Ceramiales), a bizarre red algal para- systema sexuale: cum differentiis specierum, synonymis site from the central Pacific. European Journal of auctorum, nominibus incolarum, solo locorum, tempore Phycology, 37: 269–278. florendi, offìcinalibus pharmacopoeorum. J. Nourse & C. Kuntze, O. (1891). Revisio generum plantarum. Pars II. Moran, London. Arthur Felix, Dulau & Co., U. Hoepli, Gust. A. Hurd, C.L., Harrison, P.J., Bischof, K. & Lobban, C.S. Schechert, Charles Klincksieck, Leipzig, London, Milan, (2014). Seaweed Ecology and Physiology. 2nd ed. New York, Paris. Cambridge University Press, Cambridge. Kützing, F.T. (1863). Tabulae phycologicae; oder, Abbildungen Iyengar, M. O. P., Balakrishnan, M. S. (1949). Morphology der Tange. Vol. XIII. Gedruckt auf kosten des Verfassers (in and cytology of Polysiphonia platycarpa Boergs. commission bei W. Köhne), Nordhausen. Preliminary note. Proceedings of the Indian Academy of Kützing, F.T. (1864). Tabulae phycologicae; oder, Sciences, Section B 29: 105–108. Abbildungen der Tange. Vol. XIV. Gedruckt auf kosten Iyengar, M.O.P. & Balakrishnan, M.S. (1950). Morphology and des Verfassers (in commission bei W. Köhne), cytology of Polysiphonia platycarpa Boerges. Proceedings of Nordhausen. the Indian Academy of Science B, 31:135–161. Kylin, H. (1941). Californische Rhodophyceen. Acta Karsten, U., Sawall, T., Hanelt, D., Bischof, K., Figueroa, F. Universitatis Lundensis, 37(2): 1–71. L., Flores-Moya, A. & Wiencke, C. (1998). An inventory Kylin, H. (1956). Die Gattungen der Rhodophyceen. CWK of UV-absorbing mycosporine-like amino acids in Gleerups Förlag, Lund. macroalgae from polar to warm-temperate regions. Lam, W.D., García-Fernández, M.E., Aaboal, M. & Vis, M. Botanica Marina, 41: 443–454. L. (2013). Polysiphonia subtilissima (Ceramiales, Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, Rhodophyta) from freshwater habitats in North M., Sturrock, S., Buxton, S., Cooper, A., Markowitz, S., America and Europe is confirmed as conspecific with Duran, C., Thierer, T., Ashton, B., Mentjies, P. & marine collections. Phycologia, 52: 156–160. Drummond, A. (2012). Geneious Basic: an integrated Lartillot, N., Lepage, T. & Blanquart, S. (2009). PhyloBayes and extendable desktop software platform for the 3: a Bayesian software package for phylogenetic EUROPEAN JOURNAL OF PHYCOLOGY 29

reconstruction and molecular dating. Bioinformatics, 25: Nam, K.W., Maggs, C.A., McIvor, L. & Stanhope, M.J. 2286–2288. (2000). Taxonomy and phylogeny of Osmundea Lauret, M. (1967). Morphologie, phénologie, répartition (Rhodomelaceae, Rhodophyta) in Atlantic Europe. des Polysiphonia marins du littoral Languedocien. I. Journal of Phycology, 36: 759–772. Section Oligosiphonia. Naturalia Monspeliensia, Bot., Navarro, N.P., Mansilla, A., Figueroa, F., Korbee, N., Jofre, J. 18: 347–373. & Plastino, E. (2014). Short-term effects of solar UV radia- – Lauret, M. (1970). Morphologie, phenologie, repartition tion and NO3 supply on the accumulation of mycospor- des Polysiphonia marins du littoral Languedocien. II. ine-like amino acids in Pyropia columbina (Bangiales, Section Polysiphonia. Naturalia Monspeliensia, Bot., 21: Rhodophyta) under spring ozone depletion in the sub- 121–163. Antarctic region, Chile. Botanica Marina, 57:9–20. Lee, Y. (2008). Marine Algae of Jeju. Academy Publication, Noda, M. & Kitami, T. (1971). Some species of marine Seoul. algae from Echigo Province facing the Japan Sea. Levring, T. (1941). Die Meeresalgen der Juan Fernandez- Scientific Reports Niigata University, Ser. D. (Biology), Inseln. Die Corallinaceen der Juan Fernandez-Inseln. In 8:35–52. The Natural History of Juan Fernandez and Easter Norris, R.E. (1994). Some cumophytic Rhodomelaceae Island, Vol. 2 (Skottsberg, C., editor), 601–670, (Rhodophyta) occuring in Hawaiian surf. Phycologia, 753–757. Almqvist & Wiksells Boktryckeri, Uppsala. 33: 434–443. Lüning, K. (1990). Seaweeds: Their Environment, Norris, J.N. (2014). Marine algae of the northern Gulf of Biogeography and Ecophysiology. Wiley Interscience California, II: Rhodophyta. Smithsonian Institution Publications, New York. Scholarly Press, Washington D.C. Maggs, C.A. & Hommersand, M.H. (1993). Seaweeds of the Okamura, K. (1912). Icones of Japanese Algae. Vol. II (10), British Isles. Volume 1. Rhodophyta, Part 3A Ceramiales. pls. 96–100. Privately published, Tokyo. Natural History Museum/HMSO, London. Parsons, M.J. (1975). Morphology and taxonomy of the Mamoozadeh, N.R. & Freshwater, D.W. (2011). Taxonomic Dasyaceae and the Lophothalieae (Rhodomelaceae) of notes on Caribbean Neosiphonia and Polysiphonia the Rhodophyta. Australian Journal of Botany, 23: 549– (Ceramiales, Florideophyceae): five species from 713. Florida, USA and Mexico. Botanica Marina, 54: 269– Parsons, M. J. (1980). The morphology and taxonomy of 292. Brongniartella Bory sensu Kylin (Rhodomelaceae, Mamoozadeh, N.R. & Freshwater, D.W. (2012). Rhodophyta). Phycologia, 19: 273–295. Polysiphonia sensu lato (Ceramiales, Florideophyceae) Phillips, L.E. (2001). Morphology and molecular analysis of species of Caribbean Panama including Polysiphonia the Australasian monotypic genera Lembergia and lobophoralis sp. nov. and Polysiphonia nuda sp. nov. Sonderella (Rhodomelaceae, Rhodophyta), with a Botanica Marina, 55: 317–347. description of the tribe Sonderelleae trib. nov.

Masuda, M., Kudo, T., Kawaguchi, S. & Guiry, M. D. Phycologia, 40: 487–499. (1995). Lectotypification of some marine red algae Phillips, L.E., Choi, H.G., Saunders, G.W. & Kraft, G.T. described by W. H. Harvey from Japan. Phycological (2000). The morphology, taxonomy and molecular phy- Research, 43: 191–202. logeny of Heterocladia and Trigenea (Rhodomelaceae, Masuda, M., Abe, T., Kawaguchi, S. & Phang, S. M. (2001). Rhodophyta),with delineation of the little known tribe Taxonomic notes on marine algae from Malaysia. VI. Heterocladieae. Journal of Phycology, 36: 199–219. Five species of Ceramiales (Rhodophyceae). Botanica Price, I.R. & Scott, F.J. ( 1992). The turf algal flora of the Marina, 44: 467–478. Great Barrier Reef. Part I. Rhodophyta. Townsville, McIvor, L., Maggs, C.A., Provan, J. & Stanhope, M.J. Australia: Botany Department, James Cook University. (2001). rbcL sequences reveal multiple cryptic introduc- Provan, J., Booth, D., Todd, N.P., Beatty, G.E. & Maggs, C. tions of the Japanese red alga Polysiphonia harveyi. A. (2008). Tracking biological invasions in space and Molecular Ecology, 10: 911–919. time: elucidating the invasive history of the green alga McIvor, L., Maggs, C. A., Stanhope, M. J. (2002). RbcL Codium fragile using old DNA. Diversity and sequences indicate a single evolutionary origin of multi- Distributions, 14: 343–354. nucleate cells in the red algal tribe Callithamnieae. Pueschel, C. M. (1990). Cell structure. In Biology of the Red Molecular Phylogenetics and Evolution 23: 433–446. Algae (Cole, K.M. & Sheath, R.G., editors), 7–41. Millar, A.J.K. (1990). Marine red algae of the Coffs Cambridge University Press, Cambridge. Harbour region, northern New South Wales. Rambaut, A., Suchard, M.A., Xie, W. & Drummond, A.J. Australian Systematic Botany, 3: 293–593. (2013). Tracer MCMC Trace Analysis Tool, Version Morrill, J.F. (1976). Comparative morphology and taxon- v1.6.0. Available from: http://tree.bio.ed.ac.uk/software/ omy of some dorsiventral and some parasitic tracer/ Rhodomelaceae. Thesis, University of North Carolina, Ribera, M.A. & Boudouresque, C.F. (1995). Introduced Chapel Hill. marine plants, with special reference to macroalgae: Muangmai, N., Yamagishi, Y., Maneekat, S. & mechanisms and impact. In Progress in Phycological Kaewsuralikhit, C. (2014). The new species Neosiphonia Research (Round, F.E. & Chapman, D.J., editors), 187– thailandica sp. nov. (Rhodomelaceae, Rhodophyta) from 268. Biopress Ltd. the Gulf of Thailand. Botanica Marina, 57: 459–467. Ronquist, F., Teslenko, M., Mark, P. van der, Ayres, D., Nägeli, C. (1846). Über Polysiphonia und Herposiphonia. Darling, A., Höhna, S., Larget, B., Liu, L., Suchard, M.A. Zeitschrift für wissenschaftliche Botanik, 4: 207–256. & Huelsenbeck, J.P. (2011). MrBayes 3.2: efficient Nam, K.W. & Kang, P.J. (2012). Algal flora of Korea. Bayesian phylogenetic inference and model choice across Volume 4, Number 4. Rhodophyta: Ceramiales: a large model space. Systematic Biology, 61: 539–542. Rhodomelaceae: 18 genera including Herposiphonia. Savoie, A.M. & Saunders, G.W. (2015). Evidence for the National Institute of Biological Resources, Incheon. introduction of the Asian red alga Neosiphonia japonica 30 P. DÍAZ-TAPIA ET AL.

and its introgression with Neosiphonia harveyi Stegenga, H., Bolton, J.J. & Anderson, R.J. (1997). Seaweeds (Ceramiales, Rhodophyta) in the Northwest Atlantic. of the South African west coast. Bolus Herbarium, Molecular Ecology, 24: 5917–5937. University of Cape Town, Cape Town. Savoie, A.M. & Saunders, G.W. (2016). A molecular phy- Stuercke, B. & Freshwater, D.W. (2008). Consistency of logenetic and DNA barcode assessment of the tribe morphological characters used to delimit Polysiphonia Pterosiphonieae (Ceramiales, Rhodophyta) emphasizing sensu lato species (Ceramiales, Florideophyceae): ana- the Northeast Pacific. Botany, 94: 917–939. lyses of North Carolina, USA specimens. Phycologia, Saunders, G.W. & McDevit, D.C. (2012). Methods for 47: 541–559. DNA barcoding photosynthetic protists emphasizing Stuerke, B. & Freshwater, D.W. (2010). Two new species of the macroalgae and diatoms. Methods in Molecular Polysiphonia (Ceramiales, Florideophyceae) from the Biology, 858: 207–222. western Atlantic. Botanica Marina, 53: 301–311. Saunders, G.W. & Moore, T.E. (2013). Refinements for the Suhr, J.N. von (1831). Beschreibung einiger neuen Algen. amplification and sequencing of red algal DNA barcode Flora , 14: 673–687, 709–716, 725–731. and RedToL phylogenetic markers: a summary of cur- Thiers, B. (2016) [continuously updated]. Index rent primers, profiles and strategies. Algae, 28:31–43. Herbariorum: A global directory of public herbaria and Schmitz, F. & Falkenberg, P. (1897). Rhodomelaceae. In Die associated staff. New York Botanical Garden’s Virtual natürlichen Pflanzenfamilien nebst ihren Gattungen und Herbarium. http://sweetgum.nybg.org/science/ih/. wichtigeren Arten insbesondere den Nutzpflanzen unter Thorne, J.L. & Kishino, H. (2002). Divergence time and Mitwirkung zahlreicher hervorragender Fachgelehrten, Teil evolutionary rate estimation with multilocus data. 1, Abteilung 2 (Engler, A. & Prantl, K. editors), 421–480. Systematic Biology, 51: 689–702. Verlag von Wilhelm Engelmann, Leipzig. Uwai, S. & Masuda, M. (1999). Kintarosiphonia Schneider, C. W. & Searles, R. B. (1991). Seaweeds of the (Rhodomelaceae, Ceramiales), a new red algal genus Southeastern United States: Cape Hatteras to Cape based on Pterosiphonia fibrillosa Okamura from Japan. Cañaveral. Duke University Press, Durham and London. Phycologia, 38: 225–233. Segi, T. (1949). On the systematic position of the so-called Womersley, H.B.S. (1979). Southern Australian species of Polysiphonia hakodatensis Yendo from Japan. Siebutsu, Polysiphonia Greville (Rhodophyta). Australian Journal 14: 134–139. of Botany, 27: 459–528. Segi, T. (1951). Systematic study of the genus Polysiphonia Womersley, H.B.S. (2003). The Marine Benthic Flora of from Japan and its vicinity. Journal of the Faculty of Southern Australia. Rhodophyta. Part IIID. Ceramiales- Fisheries, Prefectual University of Mie, 1: 169–272. Delesseriaceae, Sarcomeniaceae, Rhodomelaceae. Setchell, W.A. (1926). Tahitian algae collected by W.A. Australian Biological Resources Study & State Setchell, C.B. Setchell and H.E. Parks. University of Herbarium of South Australia, Canberra & Adelaide. California Publications in Botany, 12:61–142. Wulfen, F.X. (1789). Plantae rariores carinthiacae. In

Setchell, W.A. & Gardner, N.L. (1924). XXIX Expedition of Collectanea ad botanicam, chemiam, et historiam natur- the California Academy of Sciences to the Gulf of alem, spectantia cum figuris (Jacquin, J.N., editor). Ex California in 1921. The Marine Algae. Proceeding of the Officina Wappleriana, Vienna. California Academy of Science, 4: 695–949. Wynne, M.J. & Banaimoon, S.A. (1990). The occurrence of Setchell, W.A. & Gardner, N.L. (1930). Marine algae of the Jolyna laminarioides (Phaeophyta) in the Arabian Sea Revillagigedo Islands expedition in 1925. Proceedings of and the Indian Ocean and a new report of the California Academy of Sciences, 4: 109–215. Melanothamnus somalensis (Rhodophyta). Botanica Shameel, M. (1999). Melanothamnus afaqhusainii, a new Marina, 33: 213–218. red alga from the coast of Karachi. Pakistan Journal of Yang, E.U., Boo, S.M., Bhattacharya, D., Saunders, G.W., Botany, 31: 211–214. Knoll, A.H., Fredericq, S., Graf, L. & Yoon, S.H. (2016). Shavit, L., Penny, D., Hendy, M.D. & Holland, B.R. (2007). Divergence time estimates and the evolution of major The problem of rooting rapid radiations. Molecular lineages in the florideophyte red algae. Scientific Reports, Biology and Evolution, 24: 2400–2411. 6: 21361. Sherwood, A.R., Kurihara, A., Conklin, K.Y., Sauvage, T. & Yoon, H.Y. (1986). A taxonomic study of genus Presting, G.G. (2010). The Hawaiian Rhodophyta biodi- Polysiphonia (Rhodophyta) from Korea. Korean Journal versity survey (2006–2010): a summary of principal find- of Phycology, 1:3–86. ings. BMC Plant Biology, 10: 258. Yoshida, T. (1998). Marine Algae of Japan. Uchida Silva, P.C., Basson, P.W. & Moe, R.L. (1996). Catalogue of Rokakuho Publishing Co., Tokyo. the benthic marine algae of the Indian Ocean. University Zuccarello, G.C., West, J.A., Kamiya, M. & King, R.J. of California Publications in Botany, 79:1–1259. (1999). A rapid method to score plastid haplotypes in Stamatakis, A. (2014). RAxML Version 8: a tool for phylo- red seaweeds and its use in determining parental inheri- genetic analysis and post-analysis of large phylogenies. tance of plastids in the red alga (Ceramiales). Bioinformatics, 30: 1312–1313. Hydrobiologia, 401: 207–214.